Semtech Corporation Stock

Welcome to Semtech's Data Center Analyst Day 2026. Please welcome to the stage Semtech Senior Vice President, Investor Relations, Mitchell Haws.

Good morning, everyone, and welcome to Semtech's Data Center Analyst Day. We have a great program in store for you today. A lot of discussion today around where data center fits within our portfolio and how we're very well positioned for the future. We'll also talk about how that translates into our long-term financial goals. Looking at the agenda, we'll start with, sorry, Hong will kick us off with a discussion around the transformation of Semtech, where we've been on that journey and where we go from here.

Following Hong's remarks, we have the pleasure of the industry panel to talk about the future of data center networking. So, we expect a lively discussion. Following that, Imran Sherazi, our General Manager of Signal Integrity, will go into detail on the data center road map. Importantly, he'll talk about the legacy of our data center business and how that positions us for a very bright future. During lunch, if you see at the back, in addition to lunch, you'll have a chance to look at all the demos. Most of what we talk about today in terms of products will be featured in those demos. So, a good opportunity to see the technology in action.

Mark will begin at 1:00 p.m. with a discussion around the financial results for Q3. He'll talk about our long-term financial model and how that looks through the end of FY '30. Following that, we'll open the floor to questions. The senior leadership team for Semtech will all be here. So a good chance to dig deep. A couple of housekeeping items. We will make forward-looking statements today. Those statements carry certain risks and uncertainties. The risks associated with that are available today on the website and as well as our SEC filings. In addition, during the course of the discussion, we will talk about non-GAAP financial measures. There are reconciliations for those are on our website at investors.semtech.com. With that, thank you for joining us. Let's get started.

[Presentation]

Please welcome to the stage Semtech President and CEO, Dr. Hong Hou.

Good morning, everyone. Welcome to the Data Center Analyst Day. For those of you who come over from the OCP, welcome. It's very fitting for us to have the Data Center Analyst Day alongside with a great event of the OCP Global Summit. That's where the innovation was showcased. That's the direction we are heading to. Having been the CEO for Semtech slightly over 2 years. And during this time, you often ask us to hold an Analyst Day, our R&D to talk about where we differentiate and what are the acronyms of different innovations and technology.

Here we go. The reason we waited for 2 years is because we were busy in transformation. But right now, I will say, and hopefully, you see the proof that Semtech is a different company. When I visit different sites, our employees often tell me I have been working in this company for 15, 20 years, but I feel it's different very, a lot of energy over the last 2 years, our customers are telling us how did we get to this point? Today, I wanted to walk you through our journey of Semtech Rising, Semtech Transforming and Semtech Exceling.

So, let's get started. Its inflection point is happening now, and we are fundamentally changing the company's profile. With a sharper portfolio and after the divestiture announcement of a wireless module business that allow us to double down to have sharpened the focus on the core portfolio, and we have stronger growth momentum. The AI data center spending is increasing and the networking connectivity is bottleneck, and that is the area we are focusing on that is the area we're doing well. And with the increased demand, we got tremendous booking momentum. We got record level backlog. That's given us visibility for many years in the future.

So, it's the right time for us to provide a multiyear financial guidance. And the early proof shows that we are on track to build a predictable high-growth and high-margin business. I know I met many of you, if not all of you, over the last couple of years, but I still like to spend a few words on myself, I'm still as excited as leading this company at my day 1. I feel very privileged to lead the next phase of Semtech transformation and growth. I'm technical, had a technical degree in electrical engineering and probably 1 of the few CEOs has worked in 3 national labs in 3 countries: China, Japan and U.S. and also conducted research on photonics and communication at AT&T Bell Laboratories in my early days.

My industrial experience spans over communication semiconductor and data center. I've been moving around quite a bit, but that background really prepared me I feel like for this job at Semtech. The 3 products may experience the most relevant to what I do at Semtech. EMCORE, at the time I was the CEO, started as Director of Technology, and we led the high-power laser development and commercialization with supporting high-power laser support probably 80% of the broadband infrastructure for cable TV. You probably don't know that. But that's why when we had the acquisition, we know exactly what we're buying.

The second part, when I was working at Intel to lead the cloud networking group, that's where I get other connections and get the familiarity on the data center requirement. Before I step in as the CEO, I was the independent director for 1 year, that prepared me to get to know the challenges and opportunities. All of this experience translate into the time. I was able to hit the ground and running without any learning curve without a trial and error. So, I feel fortunate to come to a right place at the right time, bring the right skill set. Now a little bit more about Semtech. We got a long history. We were founded in 1960. We became a public company in 1967. Just to calibrate yourself, that's 1 year prior to Intel's founding.

So, when I engage with investors very often, I heard, "Oh, I know Semtech." My ask is which part of the history do you know about Semtech? We started defense rooted pivoted to analog and mixed signal and the acquisition in 2012 of Gennum really bringing the market-leading asset. So that's our crown jewel. We did a little bit detour in 2023 of the acquisition of Sierra Wireless, at the wrong time, which stretched our balance sheet and diluted our focus. I'm not going to draw on that, but I don't want you to pretend that never happened. So, I joined after that, and right at a time, the industry it's investing in AI infrastructure tremendously. We analyze our portfolio. We feel we have the right to win, and we are embarking a journey of transformation for Semtech.

Very established company, but it can't have the renewed vision and focus. So, today, we aspire to be the leading connectivity company of the AI era, focused on moving more data with less energy from the heart of the data center to the edge of the physical world. We calibrate our strategy, our action against this vision statement. So, we embarked our journey called Semtech Rising and 3 arcs and 1 story to transform the company. Semtech Rising to fix the fundamentals, we started that second week as I was the CEO, and we can be clear now mission accomplished for that phase. We are squarely in the Semtech Transforming that is about to build more growth engines to accelerate our business growth. And our destiny, our goal is to be the leader of the industry in what we do, that Semtech Exceling.

Again, there are 3 phases to support that vision. Let me talk about the Semtech Rising. As I mentioned about, we embarked this journey second week after I became CEO we didn't have time on luxury to do a long history study, but we know what stress does. We know what need to be done. Our positioning was not clear, and we were a mixed signal analog diverse company, that's not enough to get us focused on achieving our mission, achieving the alignment with the market opportunities. Balance sheet was stretched. To say the least, we had a leverage ratio of 10.5x at the time. And that wouldn't allow us to do anything more aggressive and then our employees, morale was low, and the engagement with the customers was lacking. So, what we did, the strategic positioning is the delineated core and noncore and put omitted resources in the core area in R&D.

And even for the core areas, we did a thorough review that we only prioritize the project that matters to the customers the most. And in the balance sheet strengthening, we did a number of transactions, and we brought the leverage ratio to slightly over 1 at this point. And we also prioritize our capital priorities in R&D and be really nimble in G&A, and we were able to go on offense as we improve our balance sheet and financial capabilities. On the customer engagement, I'm going to have a slide to talk about that in more detail. But on the employee side, we communicate often and communicated at our levels through all-hands meetings through monthly newsletter through site visits we reestablished trust and transparency. And I am very proud today, we're going to be giving out many numbers.

This number I'm most proud of, our engagement score from 2 years ago, I will hand it over with a scorecard of 57% engagement to today's 78%. That is really things matters the most. If we did not have the employees aligned and to share the clarity and the vision, we wouldn't be here today. So, with that, we were able to turn the highly leveraged company into a focused and aggressive data center pursuit. So, that to double-click on one of the fundamental culture change, it's a customer engagement process. We provide IC, the key components now including photonics to the module manufacturers and then they, in turn, to supply the integrated product to hyperscalers and xPU companies. In the past, we basically engage with the module manufacturers on this chart.

And we do not really engage much with the define, the definitional customer who designed the data center who defined their architecture. Now it's different. We engaged all levels. By leveraging our, the analog and mixed signal expertise and that is a track record and experience, we are viewed increasingly more by the hyperscalers and a technology provider as a technology partner. We can help to understand their pain points and by applying our differentiated technical analysis from simulation to purpose-built silicon to provide differentiated solutions. One, we provide that what is doable and they can define their architecture and our solution by doing this engagement is almost automatically designed in.

Of course, we engage with our direct customers very closely. This help us to accelerate the design in process help us to translate into accelerated growth. As I said, we got early proof today, we are invited to the room in the early architecture design phase by the hyperscalers, by the xPU companies. That's really a great place to be, and we are increasingly getting trusted to be the solution provider. Semtech Transforming. That's where we're squarely in this phase. The aim is to build more growth engines. Now we have the financial capability. We want to double down and intensify the R&D investment. Just to give you a ballpark number. This year, we are projecting to increase our R&D spending by 30% sequentially.

In the core area, we probably even more heavily indexed on the R&D. Why is that? To build more growth engines. Many, most of the projects, I would say, stemmed from the alignment with the customers and trying to solve their most pressing issues not by the scientific curiosity. So, we believe that the investment will inevitably develop a new product and generating new business opportunities the feedback allow us to be able to invest more. So, we are really in this good virtuous cycle to accelerate business growth. We're in a really good place. And on the portfolio optimization, we did the first divestiture of noncore assets that allow us to deliver predictable margin profile we're constantly looking at our portfolio, but we wanted to build more growth engines, data centers today, the largest 1 and we have the LoRa and we believe we can have more growth engine based on the current portfolio.

So, portfolio optimization is an ongoing journey as we increase our, expand our portfolio and expand our ambition, we will identify more gaps in capabilities and -- but M&A focus is not to increase the scale of buying revenue but more enhancing capability and apply the leverage. So, we have some ideas, and we constantly review that, that's a journey, strengthening the winning culture. Our employees get an early test of winning that is great and winning is contagious and winning is gratifying. Winning is our largest driver behind our employees. So, you're going to have a chance to meet our application engineers when we're in the break and roam around to 'see our demos.

But we are aligned in our strategic imperative growth oriented customer intimacy, innovation with differentiating solutions, do the best in whatever we do in whatever area we are in, and employee engagement. So, we'll continue to strengthen that culture, and we are well underway for transformation of the company. To show a proof point that we are positioned at inflexion and Mark is going to be talking a lot more detail on our financial projections. But here, I would just wanted to show the historic 3 bars at the trend chart. It took us a year to grow the top line about high, our mid-single -- mid-teens from the Q3 '25 to Q3 '26. But we're going to be having over 50% of growth from Q3 '26 to Q3 '27, that from the updated guidance we're going to be giving.

So, our growth is accelerating. Margin profile after divestiture of the Sierra Wireless module asset as expected in to close in Q4. Our gross margin baseline is going to be 64%, with tremendous tailwind of the mix shift into the high-margin business, and that is a floor, not a ceiling. Our earning is accelerating even more so because of the operational leverage when you have a larger scale without increasing G&A, and that would translate into more earnings per share. So, we see the early proof that the inflection is happening. It is not just a projection. This is the data. This is the fact.

Here, I said we are, today, it's not a full long Analyst Day in reviewing our product and portfolios but more data center centric is fitting to co-host with the OCP event, but also showing that in FY '24, our data center revenue only account for about 7% of total revenue. This quarter, we expect the data center revenue account for about 40% of total revenue. It's not because the other part of the portfolio is shrinking. It's just because the data center is, the growth is accelerating. We'll show you that the FY '30 forecast later. Mark is going to present that. And in that, we're going to be having the data center revenue to be about 2/3 of total revenue with a much bigger number of the total consolidated revenue.

So, we are still in the early innings of this growth, even though this market inflection points and the portfolio optimization continues, that's helped to bring the margin profile and also get a better bottom line. And our strategy is to become a leader in the data center. We may not be brought, but we want to be the best in everything we do in every product we offer. They can be copper, it can be fiber, but the theme is to provide high bandwidth, low-power solutions. We all see this chart this from BofA recently, and also this trend chart, is different maybe after this week. It's going to be revising a little bit maybe up or maybe down but you can't argue there is no, any of this level of investment is unprecedented, and we are sitting in the front and center of this in data center the gating items used to be compute capabilities.

Today, when we engage with the customers, we were often told you, the networking connectivity, it's a bottleneck. We have the opportunity to unleash almost as much as 2x of the compute power. And what the customer wants, it's a high bandwidth low power, low latency and high bandwidth density solutions. That's exactly what we do. So, that's what we play in data center. We don't do cooling or power delivery. We don't do compute. We don't do memory either, but we're just laser focused on providing the best network connectivity solutions.

So, talking about connectivity, you heard the terminology many times, the different connectivity, but they all share the same requirement again, high bandwidth, lower power consumption and high bandwidth density and low latency, starting from scale across. This is the so-called in the past, telecom, metro, but now it's proliferating into more, on the same campus between different buildings. The reach is 2 kilometers and above. Through Gennum acquisition later on, I will discuss in more detail, we offer the industry-leading gain chips to be integrated in tunable lasers and to be the high-end laser for metro and our scale-across applications. Our laser is ready for coherent lite application as it's now, and we are engaging with multiple customers on that.

So, we have some play in scale across that's only going to expand what is most of the ramp we're experiencing is a scale out. So, that is for interconnecting the Ethernet switch fabric in data center. And our solutions is incorporated in the fully retimed, optical pluggable modules, LRO, LPO, AOC, you name it. A lot of acronyms and Imran is going to be doing justice in his presentation to explain in more detail and lift up the hood and show what we have in there, what's our content in there. We're supporting 800 gig. We're ramping 1.6T with designing 3.2T. We provide multi-generational solutions co-designing that with our customers.

In front of the center, that is our most of the activity at this point. Good thing is volume compared to scale across estimated. You tell me a 15, 20, 25x of the volume is changing every day. Scale-in, scale up first, that is even higher volume, 10x more. Scale up is to interconnect the servers, the switches and at the rack level and cross racks within a cluster. The scale-up we provide solutions both copper and fiber, copper through a redriver or linear equalizer and when it's integrated in a cable the ACC is a great backplane interconnect solution, and that business is ramping with key hyperscalers.

There's a lot of design activities of linear equalizer integrated on the board because later on, when Imran presents, you will see that signal integrity is going to be compromised when you transmit high data rate across the PCBs, the most elegant way to solve that problem is integrate redrivers, some use retimers a little bit penalty on power side, but you'll be able to clean up the signal and allow the signal to reach the other end without a distortion. Use copper when you can, use fiber when you need to. That has been our guiding principle and validated by Jensen in the GTC meeting.

Otherwise, many of the analysts was asking me for a terminal value of copper. Don't put the terminal value yet. Copper is going to be playing in 224 gig racks. Copper is going to be even more playing in the 448 gig racks. I mentioned about a 3.2T design activities that's replicable. The optics on board is tremendous activities in there. They see this as increasingly a very viable solution. Then NPO, the CPO, the optical interposers, I assume those are really the key demos got to get a chance to walk on the floor yesterday for a couple of hours at OCP. I probably already counted 15, 20 different demos of the NPOs. Our solution is there. If not there today, we're tweaking the form factor and it's going to be incorporated tomorrow. The reason for that is current NPO design is 200 gig per lane.

That is an IP that performance is already validated through the pluggable transceivers. NPO, in a way, is LPO on steroids; it is a highly packed form factor of LPO performance is already validated, and we are in the front end center in providing solutions in there. So that is a market moving toward us. Market is moving toward analog, and we do nothing but analog solutions at this point. Scale in a lot of discussions and people really more forward-looking. And when you run into the connectivity wall and power wall and you can use interconnect between different chips with the chiplets or optical interposer. So, we are closely in tune with our customers on the different solutions. So, that is a connectivity topology.

And as the data rate increases, just by providing the best PMD, or physical media devices, TIA and driver, it's not sufficient because people will give you the spec with the broadband. We don't have a lot of performance margin to give out. So, that's why we need to expand our portfolio to be able to have optical components alongside with our electronic components. This co optimization will allow us to provide a best chipset solution to the customers with reference design. Fitting in a theme of building additional engine, we did a couple of acquisitions of photonic assets. In March, we announced our acquisition of HieFo is a company going through different recognitions, but rooted with Caltech, Ortel, Lucent in 2000, spent $3 billion buy the company and EMCORE when I was there, and we bought this company from Lucent and subsequently, we build high-power lasers.

And we leveraged 46 years of the heritage, and we know that we are leaders in the optical component in providing the best our conversion efficiency the design allow us to provide the best beam profile allowed to low loss coupling into the fiber, our waveguide and also the noise figures. So, this is really a great asset and grid extension. We have vertically integrated operating the fab in L.A. area from MOCVD epi to fab processing to test and chip-on-carrier packaging, we can utilize the entire ecosystem to increase our capacity. Speaking of capacity, today, we get more demand than we have the capacity. So, we are in the process of increasing capacity. We project to increase capacity by 3, 4x by the end of this year and to serve the customers later on in different applications.

One other way to increase our capacity is that we did another acquisition of photonic asset in June of this year, D-Tech. So, we brought on board the world-class designers for high-speed photodiode. And the acquisition also allow us to have the immediate access to 7,000 square feet clean room facility. So, we got preown the equipment. We're moving into this space. That's how we increase the capacity by 3, 4x end of this year. There is a little different facility, but within driving distance of 10, 12 miles. And we are also focusing on design photodiodes and project to provide samples and demo in the upcoming OFC.

So, our photonics portfolio right now is still limited, but we are not planning to build everything for everyone. We want it to be answer ourselves on the strength and really proliferate strategically. For example, on laser side, I mentioned about our game chip is already class-leading provide the highest power in the tunable lasers. We have a solid customer base with now HieFo inside of Semtech for 6 months, we are able to increase our customer base pretty significantly. As I mentioned already, our lasers in the same design already provide low noise, high-power and narrow linewidth which is our ideal for coherent and light optical transceivers.

We're engaging with multiple module manufacturers and then having our samples in their evaluation and qualification queue. Some of the early results indicating that we're better than any incumbent lasers they are using right now. The large market opportunity, the industry is still experiencing a lot of shortages is for IMDD, intensity modulated direct detect applications. You're hearing the laser shortages. Of course, we got a lot of interest in that area. But we are, we don't have a lot of capacity to serve them. Instead, we use our differentiation. We use very high power lasers delivering power, the 200 to 250-milliwatt power CW lasers, which is different from, say, a 30-milliwatt CW lasers, you have to use 1 laser per channel and lasers in optical transceivers.

Now with our UHP CW lasers, you can have 1 laser to drive 8 channel and silicon photonics modulators, but we have to engage selectively because we wanted to engage with the customers, they really value the U.S. made and for security, for the supply chain surety. And we use this as a platform to design for a future and develop OCI, CPO scale-up players working with the partners. We provide UHP 300, 400-milliwatt lasers with the semiconductor optical amplifier together with their Wavelength definition capability, we're able to provide DWDM lasers at high power for CPO scale up. Again, we're not trying to be everything for everyone.

In Photonics, we are building strength outward and with the customer pulling us forward. The portfolio expansion can translate directly into the content increase in our business. So, here, it's a chart. Probably you can take a picture, you can remember that I mentioned about the 800 gig, our content is single -- high single digit. When the industry evolving into 3.2T, for example, we'll be able to monetizing more with the same component functionality with more complexity like a TIA and drivers plus adding more content like CW lasers and photodiode and even modulators. This is really very important because the most of the architecture shift in the future trend, increase our content.

That's why we're saying the market is moving toward us rather than away from us. So, right now, the content on 800 gig and through LPO increases through NPO increase even more, and we are getting ready for the components to be inserted in the 3.2T generation. Okay. About the market size. So, we use 650 Group, and we'll have the founder and CEO of the 650 Group, Alan Weckel, is going to be on the panel later on. So, we use their data, and they are most in tune for the networking side. If you look at the slides in the teal color, that's the size of PMD, the TIA and drivers, by the way, we adjust it only for our product, not the entire TAM. And of course, this does not include any additional opportunities that we're evaluating from M&A point of view. But that incorporate what's in, on our road map for organic expansion opportunities.

Lasers and game chips and SOAs, the addition of photonic components increase of SAM pretty significantly. I still feel that active copper have more opportunity than the market giving credit for because when you are going to a higher data rate, you just don't have other ways to get to the other end without redriver or retimer. Either way, so in 4 years, you see the SAM can increase by 5x. That's tremendous. That is a great opportunity. For us, we grew revenue through avenues. We can simply write with the wave and the size and trend will benefit us. That's #1. The #2, the content increase. I mentioned about 10x potential from the previous slide. #3, we aspire to continue our journey like we have demonstrated on 800 gig TIA gain market share on future generations. So that gives us a more growth driver as well.

So, what give us the right to win? Why Semtech? I mentioned about a technical leadership and Imran is going to be talking more eloquently about that later on. I'm not going to dwell on that. We have the track record. We have the data rather than proposal, customer intimacy this engagement model works, and we'll continue to gain customer trust, and we want it to be also easy to work with the customers. This is a sentiment from the customer unprompt. It's not what I wanted to do, and we really engage with the customers without arrogance and we try to be problem solving.

Operations excellence. So, we have started to planning process with our customers on a monthly basis. Now we proliferate that number to over 20. That gives us an insight not only for what immediately they need and also what they're planning in the future. So, that allows us to proactively design for capacity. And on Tuesday, we announced a multiyear strategic agreement with our great strategic partner, Tower Semiconductor. That agreement provides secure capacity to support our business growth in the horizon we are going to be providing a forecast. So, we are also the hallmark of the operations for Semtech team is to provide operations excellence all the time. So those are the 3 pillars to support our vision.

I will be doing a disservice if I don't mention about spend a minute to talk about the LoRa because that's our second growth driver. And we also did some realignment with the market on the LoRa strategy. We focus on silicon develop the best capability, lowest power, longest reach and higher bandwidth and provide market enablement. So, now we have the dual band to provide enhanced bandwidth. We have the LoRa provide enhanced the capability on another protocol we have Amazon to address the mass market of consumer side through Sidewalk, we have been traditionally been very strong in providing robust connectivity solutions for medical devices the enablement through innovation forums through software development stack through the webinar and customer engagement to list the entire ecosystem.

So, they work very hard alongside with us, that allow us to proliferate the application verticals from traditional 3 to 4 to now 8, 9 and 10 or even more drilling. That is a great growth driver. Early on, I mentioned that we have planned to review the rest of our portfolio to continue to develop more differentiation, continue to develop better drivers. So, what does that translate to? So, Mark is going to be talking about this in a lot more detail. We're projecting by FY '30, our revenue is $3.2 billion, mostly driven by data center and LoRa, that translates into the consolidated growth rate of 38% compound. Just get it translated into the multiples, 75% CAGR translating to 9x growth in 4 years in revenue. 30% CAGR translate into nearly 3x of growth of LoRa revenue from, in 4 years.

And then consolidated is going to be about 3.5x. I wanted to provide a little bit more color on this projection and we have seen a lot of project in the hockey stick. But I wanted to stress that we heavily indexed on the near term because we have the visibility and the confidence and conviction and so the FY '27 in FY '28, later on, Mark is going to be providing, for example, data center side, we project over 100% year-over-year growth. The rest of the horizon we use the market CAGR. But as I said, we aspire to provide a 50% higher growth rate than the market. So, as for the bottom line, Mark is going to be discussing that in more detail. And now let me bring back our journey to our destination.

Semtech Rising, done. Semtech Transforming, we're in progress is accelerating. Where do we want to bring the company to? Semtech Exceling, wanted to be providing higher growth than the market, better profitability than the market to customers. We aspire to be their partner of their choice to employees. We want them to be energized, motivated into investors. We thank you for the trust you place in us and in our team and we want you to be handsomely rewarded. The last slide, the takeaway Semtech is fundamentally different companies today. I'm not going to go through every bullet point. And hopefully, you see the data, again, is not the aspiration and we are pivoting in this really, we are inflecting and this is really in the early innings.

Let me close on a personal note. When I started as the CEO a couple of years ago, I ask our employees and some of our investors in the audience, give me the benefit of doubt. I will not let you down. Hopefully, you see that the early proof we delivered some results. But to me, to my entire team to the Board, this is just the beginning. The best of Semtech is yet to come. Thank you very much. So, next on the agenda is the industry panel, is the AI connectivity imperatives. So I am, it's my pleasure to introduce our moderator, Mark Edelstone in the company's trusted adviser and also the Chairman of Morgan Stanley Global Semiconductor banking. Mark?

Good morning, everybody. Never underestimate the importance of leadership. That is one of my fun things. I've been doing this for a very long time. And to see Semtech today is quite impressive, quite amazing. Hong in his earlier comments said that knowing Semtech is a question of what period you're talking about. I've been doing this since 1989. And so I can vividly remember my discussions with Jack Poe, who is the CEO here from 1985 to 2003. And the difference in this company today under Hong's leadership is impressive, amazing, and it's great to see. This company is in a great spot. It's an honor on to be able to do this for you today. We've got a great panel. I'll have them come in a minute, but we're going to be able in the next hour to provide, I think, a pretty good perspective on what's going on in the market.

Many of these things you'll know. Other things I hope it will be a deeper dive for you as well. But Hong and his team have actually put together some real subject matter experts around this topic, starting with Alan Weckel. Alan is the Founder of 650 Group. He's basically been doing research for a couple of decades. He's been running this company since 2017. David Ramku is the Head of Infrastructure at Meta. So we'll get a pretty good perspective of what's going on with those folks that are spending lots of money on AI infra to deliver what's happening going forward in all the AI workloads.

And then Alexis Bjorlin, who is deep technologist, such matter expert on this as well, really great background. Currently now the Chief Strategy Officer at General Catalyst which is a well-known VC investing in this sector. So, I'd like then, of course, Hong as part of the panel as well. So I'd like the 3 of them to please come up and join me on stage. Okay. Thank you so much. So, what I'd love to do, I gave just a brief background of the 3 of you, but it'd be helpful, I think, you're such domain experts in things, and we're going to cover things from the market technology all those kinds of things. So, it might be helpful to just quickly give a little bit more detail on your background.

What are the things that you've done that make you impactful to basically help share inputs to this audience. And then as you do that, also, the first question you could just do at the same time basically would be to just talk about the single biggest change you expect to see in data center networking here in the next 5 years. So maybe starting with Alan, and then we'll just go down the list, and we'll go from there.

Yes. And thanks so much for having me. So, I've been in research for over 20 years. Prior to that, I did network design at Raytheon. So, my entire career has been in networking and recently in data center. And we track over 40 different markets from compute storage networking to semiconductors. So, we're kind of at the end of the day, unit counters when all is said and done out there. And the amount of change is just incredible, right? I think the way I think about this from a networking perspective is we're going through a great transformation in terms of speeds and feeds, but fundamentally, we're increasing the content as we go.

So, networking is kind of transforming twice, both from a higher speed, higher bandwidth perspective and then also significantly more content. And that ultimately drives the networking market from tens of billions to hundreds of billions and then eventually, right, to that kind of magical trillion market just for networking itself.

And the biggest change?

The biggest change is the pace of innovation, right? If we think about what's occurring in networking, we're going from 1 network to 4 networks in order to support AI that's something that historically we've never seen at an industry level before.

Okay. David?

I'm David Ramku. I'm a part of a Meta infrastructure organization. My team is responsible for technical program management, product management, ecosystem and partnerships engagement. We also kind of lead the internal prioritization of our programs. We're part of the infrastructure team that's responsible for the physical hardware that we place in our data centers, and that ranges from service and racks that we design that are based on merchant silicon and every name you can think of is obviously represented there. We also spend a fair bit of effort on building our own silicon for multiple applications, including AI accelerators, network products, et cetera. Prior to joining Meta about 4.5 years ago, I spent 18 years at Intel. My last role there was leading the infrastructure processing unit business within the Intel Connectivity group.

And biggest change coming?

Biggest change coming. How does the saying go predictions are hard, especially about the future. Probably the network evolves from being the plumbing of the data center to being the enabler of compute architectures that unlock new possibilities for our AI applications.

Alexis?

Excellent. So, thank you for having me here. So, I currently serve as the Chief Strategy Officer for General Catalyst, but have a deep career spanning NVIDIA, Meta, Broadcom and Intel. And before that, some really excellent Source Photonics in the optics and networking space, my entire career. So, sorry about that. So, I think I think one of the most important elements that we'll see coming down the pipe for disruptive change, of course, is co-packaged optics. Optics closer and closer to the silicon, whether that manifests itself in co-packers optics, near package optics, what we're seeing is an unlock of bandwidth that's required to enable next-generation architectures, as David's saying.

So, I think that's the big disruptive change that's coming up. But when we think about the evolving landscape of compute and memory architectures, it's all still bottlenecked and constrained by the network, and the network becomes the enabler to high-performance heterogeneous compute systems in the future.

Fantastic. All right. So, and everybody knows Hong already, and he's covered his materials. So, obviously, there's lots of amazing things happening in the sector. I think it's pretty clear that what's happening in networking in the data center is going to be very impactful in terms of how the rest of this whole AI deployment is going to play out. And I think we all know we're very, very early here. I mean, ChatGPT was only 4 years ago as an example. So, tip of the iceberg, in my opinion, despite what we're seeing in the marketplace today, and we'll talk maybe a little bit more about that, too.

But just to set the stage a little bit. So, Alan, you just said $1 trillion opportunity in networking over time. And we know that this year, the hyperscale is just spending trillion dollars plus or minus. Just give us your sense of how you think it breaks out on the networking piece between scale out, scale across scale up? And then what do you see as the biggest, the fastest growers within that?

Yes. So, scale-out is the largest domain size today. That's the first market that's going to cross that $100 billion threshold. In the longer term, scale-up will become the larger market therefore, is the fastest growing. But that does not take anything away from scale across, right? We're still in the early days of locking scale across, whether that means building to building in a very large campus or connecting these campuses together. This is something that historically the hyperscalers didn't have to worry about in the volumes that we have.

But kind of agentic is unlocking that agents are by definition chatty and trying to engage with resources that sit in other locations. So, as we move from training to inference, we really do get the unlock and scale across and that will have a kind of growth rate. Each of these is a very healthy market, and you kind of add them up, and that's where you get the significant TAM expansion. Fundamentally, networking is becoming a larger part of CapEx as time goes forward.

Okay. Fantastic. David, you're in a unique position here representing, obviously, what's happening in the players that are putting forth all the infrastructure and things not specific to Meta because I recognize that puts you in a sensitive spot. But just kind of give your perspective as someone who is trying to put in all this infrastructure, what are the sort of key things that you're focused on relative to seeing payback from the types of things that are required to basically make this all happen for the rest of the workloads that people are using.

So, if you're asking in the context of how we think about kind of setting our budgets and kind of projecting out our infrastructure, obviously, we look at internal demand, and we look at how the demand plays out over the years. The, obviously, demand is plentiful. So, the next key factor becomes power availability. And that generally becomes the gating factor of how much infrastructure you can add. Obviously, we look at TCO for the various solutions that we're implementing. We also try to map out the demand shape on training versus inference, as our models get bigger, we need larger inference clusters.

On cluster size, you're trading, you're generally training time versus the cluster size. And the last, but obviously, it's become really important. Supply availability is really important. So, as we look out in time, do we think that there is supply available for us to build to our ambitions. So, to kind of recap, we think about power relative to our demand, we think of our demand shape and then we think of supply assurance and availability. There are some investments we make that happen well ahead of the actual NTI program that we deploy.

So we know that we have to invest in capabilities before that actually intercept the program, optics being an example of that. We've invested in optics, and we've done Optics research for a pretty long period in anticipation that when a product comes that intercepts, it gets integrated with that program. Otherwise, the duration of those deployments would get much longer.

Great. So, maybe following on that thread a little bit of Alexis. You've got amazing experience, you've worked for the biggest and baddest companies effectively in AI. Your most recent was DGX for NVIDIA, but before that, Meta, Broadcom, so you've seen a lot put yourselves in the shoes of someone who has to make these decisions around ROI. How do you see that today, recognizing that there's obviously a whole lot of front-loading that needs to take place, as David just talked about, how should investors think about the ROI on these investments today as we're thinking about the monetization of tokens and all the things that we're going to get out of AI?

Yes. Obviously, I think the world is trying to figure out how to think about this. But the current the current way that we look at this is the return on investment of deploying AI infrastructure really is looking at the monetization models from. Currently, we moved away from simple metrics of dollars per GP per hour, $2 dollars per token. And I think where the world will find itself is in dollars per outcome. So, one of the most important things to focus on is AI diffusion into the enterprises into sovereign applications and understanding the outcomes that we can unlock with AI and those outcomes becoming value that we're able to attribute down to the investments.

We're not there fully yet because we haven't been able to map out that full value chain. But what we are seeing is a tremendous rise in the use of AI in the use of tokens in the balance between the proprietary closed-source models and the advent of open source models that are achieving very, almost, they're in rapid pace to achieve similar outcomes. So, as we see the enterprise is deploying a mixture of proprietary and open models what it affords enterprises more choice in the cost structure of their deployments. So when we think about the deployment of more and more AI infrastructure, I'm fairly bullish on that. So, I think people are looking at this in terms, first of all, just in the infrastructure itself as a capital return.

So, you look at it in the land power shell, you look at how do we achieve an unlock of the bottlenecks that we have today. The model neck currently, everybody is looking at is the access to powered land to power itself so that we can scale with the ambition of the industry. And then when we look at an ROI at every different level of this, it's land powered shell, typically, people are looking for a 15% IRR. How do you bring that together? How do you execute that?

When you're looking at deploying the AI infrastructure itself, it really moves up the stack. So, I think we're going to see a tremendous value shift across the chain, where value is accruing in certain parts of this today. It used to be accruing in the model and application and in the semiconductors. We're going to see that value shift across the chain as this becomes as we continue to deploy at massive numbers of gigawatt scale.

And maybe just as a follow-up. I know you're relatively new in your role at General Catalyst. But as the Chief Strategy Officer, presumably, you're thinking very hard about dollars that are going to be deployed by your funds. And so what's your observation as you look at it and look at the math behind it, do the IRRs look like they are going to be achieved based on the opportunities that you see across the AI infra opportunity set?

Yes. Yes, I think that there is a compelling case for the value to be achieved for, as I said, at each layer of the stack, there's a different math. There's a different algorithm behind it. But one thing that is for sure is that the importance of compute and compute infrastructure as we see this evolution, whether the evolution as anything, this is a rising tide, and we'll see minor corrections back and forth, but the trend that's there to the ability for AI to deliver significant value and outcomes, we believe it's going to be across the entire stack.

Great. For you, Hong, and for those in the audience, I sat in those seats for a very long period of time as a research analyst. I did that for 18 years. I've now been a banker in this sector for 19. So, kind of 37 years of seeing how things evolve. And at Morgan Stanley, we've looked at all these cycles. You've probably seen research that we've done over the years at every generation of compute, starting with mainframes, going to minis, going to PCs, client server, wired Internet, mobile cloud, all those things. It's typically rest in about a decade and it's been an order of magnitude increase in terms of units, demand, all those kinds of things, returns.

And now here in this AI era, it feels to me that it's well more than an order of magnitude coming. You laid out Hong, a very robust view of what the world looks like in 2030. I agree with it. I bet you're probably low on those numbers is the Mark Edelstone point of view. But we all recognize that things can happen beyond our control. And I know that as a leader here for this company, you, Mark and us, your team, are very focused on under-promising and over-delivering and making sure that you're looking for inflection points where things could change.

So, with that being said, what are the sort of the near-term demand signals that you're looking at to make sure that things are on track and making sure you can deliver when you look out just over a shorter period of time, like over the next few quarters as an example?

Yes. Thank you, Mark. Yes. So, we provide a multiyear guidance. And as I said, the near term, we have much better visibility the booking momentum has continued to accelerate. Over the last 4 quarters, we have been reaching the record level of backlog every quarter. And while also triangulate and compare notes with our industry peers through, say, Morgan Stanley event and other industrial events. What we see that share the same sentiment. In the near term, for the next 2 years, the demand horizon is very solid and there is no single customer coming in at hedging and say, "Well, can we dial down a little bit.?" So, I get more cause these days to now help us to accelerate they pull in some of the demand, even though we are scheduled for Q2 next year, and we wanted to have it earlier.

And there are no sense we sensed that the industry, the channel is stacking with the inventories. So, that gives us a conviction and confidence for the near term. Then longer term, certainly, right now, the market is a little choppy. They look at all the different things. But either way, is unprecedented investment cycle and a great opportunity. At the AI user, we know that today, I would say the company-wise, we don't really know how to work without AI.

We're using AI in every area in our design, market analysis in our node conversion in our disposition of material yield analysis failure analysis, everything, in every area. So, that gives us the confidence. We are increasing the investment spending on the side. And I think we asked ourselves in the beginning, what is the return on investment. But now the baseline is with AI, we don't ask that question anymore. Simply put, we don't really know, it's not going to be coming, going back to the, the days without AI.

Yes. And it feels to me as well that just this opportunity here is so enormous as we just have to completely change the way that compute architectures have been done in the past, Jensen talks about AI factories and so on. And it's just gating factors here that prevent in this cycle things getting overheated versus what we've seen in prior cycles. So, land shell power, wafers, there are just some natural gating things that prevent access is getting built up, which, to me, if demand holds and I think we're still early on in the applications, one would think that it should go for a while.

One of the things though, Alan, that we hear from people is just, the overall cash flows, in some respects, people could argue the world this turned upside down. I love the way the world works today because all the companies that I think have just done more to transform our world in technology, being semiconductor companies now have pricing power. They've got variable cost business models by and large. And a lot of the CapEx burden is now being undertaken by their customers.

And they're obviously powerful, like David's company and others. But what's your view of just how long that can last? And it sort of comes back a little bit to Alexis's point of view on what the ROI is on those investments. But what's your just high-level view of how long can we see this? Or are there issues that we should be focused on relative to hyperscalers, neo clouds, enterprise all the spend that's necessary to deliver the goodness that's coming down the road?

Yes. I think it's very sustainable. So, today, we could say this is focused more on hyperscaler space or hyperscaler plus neo clouds. And what we're seeing on the early days of inference and Agentic is an incredible unlock. I can even mention that like internally to themselves, right? But the big opportunity here is going to be the enterprise unlock shortly after that, like the physical AI unlock. And those are kind of tremendous cycles. And if you look at the amount of investment relative to the unlock or relative to GDP or cash flow at a company level, we're still very much in the early stages of that. And I think that early stage plus sustainability plus what you mentioned, right, some natural gatekeeping items in terms of growth. allowed us to be a multiyear or, say, decade plus cycle to it.

Fantastic. Maybe on that point then, Alexis, if we see any type of slowdown in just the cadence of models and creation and how that evolves. Does that you think caused things to slow down for networking? Or is inference so powerful that it's just going to carry the day?

I think we're seeing a massive conversion from data center and AI infrastructure being used from training to inference. And what we see working in the space today or what I personally have visibility to is an insatiable demand for compute for inference. So, I think even if the models were to slow down, we are just at the very beginning of this curve of people adopting AI and leveraging the current model fine-tuning the models fine-tuning them on their own custom data. We're starting to see real unlocks. And the beauty of it is, as the deployments become more efficient, we're seeing even more demand.

So, it is that Jevons paradox of once people get access to the right model, they distill it on their own data, all of a sudden, they're unlocking higher value and being able to deploy further. I mean, I really believe we're just at the beginning of this. So, I don't think that, that model slowdown would have an impact on the insatiable need for infrastructure.

Yes. Agreed. So, David, maybe on that point, because I would fully agree with what Alexis just said, how does that impact the way that you're basically building the network going forward from an inference demand perspective?

When we think of the requirements of training versus inference. Training requires a larger cluster that is more synchronous in its bandwidth requirements. Inference is a bit more bursty, capacity can be a bit less predictable. You end up focusing more when it comes to inference more on scale up versus scale out. The other thing we think through is the larger context windows, which people get accustomed to appreciating require greater memory bandwidth and the agentic workloads, obviously require a tighter coupling of GPU and CPU. So, as we think about the capacity we put in place, we think about those inference requirements relative to trading.

Great. Maybe for you, Hong, then, what's your sort of view? What do you see as being unique here about the networking block effectively compared with compute and memory and how does that, how does it matter? And where's the value creation that's going to come from here over time?

Yes, I'm sorry, I couldn't, can you repeat your question?

Yes, sure. It's really the question, what's unique about the networking block compared with compute and memory that's going to create the value around the networking opportunity that exists?

Right. That's -- so as I mentioned, when we understand and we often get asked by the customers that the compute power is plentiful in there and is innovating in an accelerated pace. Every 12 to 18 months, there's a new generation of processors coming out. But we do not in the connectivity space, innovate as fast. So, we've become the bottleneck. And power is another key consideration. And every time the current connectivity infrastructure rely on DSPs, it almost feels like there's a crutches you have to use. If you can design a clean signal path, you have DSP to do the cleanup job.

But what we try to do is using an analog solution to provide a clear path for it. I do believe we can provide more value in the entire data center by innovating and providing a better solution in connectivity. I think we're getting a lot of encouragement and pressure and become accelerated innovation pace because we are the choke point of the entire stack to monetize, to provide more value. So, we can be enabling factor for the future.

Yes. Maybe when you think about that, Hong, as just a follow-up, it feels to me like there are so many different companies, the large ones as well as, obviously, lots of start-ups that are focused more on the compute side of things. And yet on this whole networking connectivity that's going to create a lot more, a lot of value to make sure that we can break through the copper wall and through the memory wall, all those things. that the number of players that have the real core competencies to do this and what Semtech can do, feels like a pretty small group of people. So, maybe just talk a little bit about that. What you see is the real core competencies and the competitive dynamics around that vis-a-vis Semtech?

Yes. So, that's a very good point. You look at the entire ecosystem, there seem to be more people work on the compute side and the networking side I would say people start, the ecosystems just start realizing the criticalness of the connectivity. For example, Alexis once she was at Intel which is back in 2016, she launched the effort of the CPO. At that time, we did a demo in the common comments is that, "Oh, we don't need that much power. We don't need that much bandwidth." So, even though we had a technology, but it didn't get commercialized until the compute really crank up so much data for AI computing and bandwidth density need to be increased dramatically.

Now all of a sudden, that becomes a trend. I think the connectivity in a way is always, at the entire time, we call adjacent, it's not a front and center. The front end center is still the compute. So, that's probably part of the reason the entire ecosystem seems to be focusing on the compute more than connectivity and that analog is kind of like an old-fashioned way. And so now it's in the front stage because they recognize the power advantage. For example, the LPO provide power saving by 90% compared to retimer.

And people did not believe it can and said, "Wow, you really, why?" mean now they start recognizing that. But you see more innovative companies and starting up. And I think the market is going to be driving the ecosystem to innovate. They're talking about the links by DLB lasers, by micro VCSEL arrays, micro LEDs and even the RF. So people start paying more attention to the connectivity at this point.

I think it's a really great point, Hong, because we talk about data centers. I mean, in the world that I'm living in where everything is measured in power. It's how many megawatts and then you back calculate from that, how many accelerators, how many CPUs you could fit into that power envelope. And so we've seen the switch from simply model hardware codesign. It used to be hardware software co-design to model hardware, codesign model, silicon. And now it's really, we're optimizing the entire system around power. And so some of these, the analog approaches and anything that could save power or where we're innovative in that network.

And then how do you build the orchestration layer of the software to be able to take into account the underlying layout of the compute system, the network, the network bandwidth, it's all being co-optimized. So, it's a tremendous opportunity. When we see a constraint like power, it's really truly grid-connected power where now everybody is building all the behind-the-meter power solutions just to unlock the capabilities. I mean innovation is happening everywhere. And ideas that didn't meet the ROI for investment before are now being looked at again. So, it's a renaissance across the network, I would say.

Yes. David, would you add anything to that relative to, we've kind of laid out where some of the constraints are? Obviously, a lot of innovation is going to happen to deal with those. But thinking about power rack density optics, anything else you'd add to what you just heard?

Yes. I mean if I think of the network power relative to everything else we use is probably single-digit percentage number. Now obviously, we're deploying gigawatts of total capacity. So, single-digit ends up mattering but the real value of the power, of the benefit of optics, in my view is from the architectural innovations that it enables. So, for example, I'm somewhat, if you guys had a chance to walk around the OCP floor, you've got to see that some of the racks now are getting a little bit bigger, but that only allows you to extend the scale-up domain so much.

If you want to get to a scale-up domain of, let's say, 576 GPUs, you need to find a way to connect 576 GPUs together. It's, and for them to act as 1 optics is what makes it possible. So, I think of optics as a lot of the value delivers is the added performance that those larger scale up domains, as an example, enabled. That wouldn't otherwise be possible or at the campus level at the cluster level, we can now, optics allows us to have these large clusters that are pretty significant distances or a lot more distant from each other than they used to be. Our, so I think of the value of optics significantly higher in that front.

Our Prometheus cluster, in New Albany, Ohio, where we trained our latest model, Muse Wonder. You've got buildings that are a couple of kilometers away. You've got buildings that are as much as 60 miles away that we're adding to that campus. It's a gigawatt cluster, and optics allowed us to make that possible, make that act as 1. Obviously, we had to put a lot of innovation in place to allow for that. We build a back-end aggregator site, which is basically a super spine.

It allows us to aggregate regions of 20,000 or so GPUs into 1. We have to build deep buffer so that traffic that's coming in from a couple of miles away doesn't get dropped. It gets kind of merged and inserted with everything else. But again, like in my view, the power is important. But the architectural capabilities and the performance you get from those is where a ton of the value comes from.

I'm glad to see there's just no complexity whatsoever in your business and the things you need to make decisions on. But maybe as you think about the things that are quite important to you or to other operators as well, how should people think about what to build internally and own internally as your real value prop versus what should come from the merchant market?

Yes, that's, we basically think through about kind of 3 high-level questions if we decide to build versus buy. Do we have the ability to optimize the full stack, hardware, software compiler if I build something internally and can I do that better? Is the volume sufficient to justify the NRE because these aren't small investments that we have to make? And then is there a TCO advantage as I look to scale out this internal solution?

Generally, we've taken the approach, though, that even when all those, all 3 of those questions are positive, we still want the fleet level to maintain a portfolio that includes both merchant silicon solutions as well as our own internal silicon solutions. The reason being is we think that fleet allows for a more resilient data center, and it allows us for better options as we scale our solutions. So, but generally, those are kind of the 3 high-level questions on how we think on make versus buy.

Great. Maybe Alan, coming back to you because we just talked a little bit about power and obviously, that's a big issue, and people now are talking about compute as a function of power. And the numbers I keep hearing just keep changing, it seems like every quarter, every year in terms of like what is the amount of spend on a capacity effectively. So, from the research that you're doing, what do you think that is today? And does that change in the future? Because I keep every time I hear someone tell me sitting in David set or someone else's see, it's a pretty high number. So, just sort of cost per gigawatt to deploy.

Yes. So, I think that's really operator-specific, right? It's a very different answer if you're a GPU as a service versus more of like a search or social platform and what you're trying to do with those workloads. But networking right now is kind of in that, we'll call it, 12% to 15% range. And that's going to go towards the 20% range in terms of spend per gigawatt and ultimately, networking is that tool that unlocks everything that we said like networking is a tool that unlocks power. It allows you to go that additional power source. Networking is a tool that allows you to unlock context windows, right? Pull that memory together. So, networking is a required tool in all of these unlocks. It's like the only universal tool that lets us solve all these problems.

But just so we get the denominator right, what's the, what's your view of total dollars per gigawatt all-in, compute, everything that's going in to basically make that happen.

Yes. So, total dollars per gigawatt is somewhere around $35 billion to $50 billion depending on what you're doing, build versus buy, et cetera. And I would expect that to increase with time, not decrease with time.

Yes. I just heard someone yesterday tell me $80 billion. Just, it's like the numbers just keep going up. And obviously, that's in part because complexity is going up as well, and this stuff is not inexpensive for sure either. Okay. So, Hong, kind of your perspective of proprietary versus open networks. How should we think about the opportunity set of what things are going to be just open standards versus things that might be more proprietary as it relates to the networking piece?

Yes. So, it is great that right now, I think we are doing both the standards and the purpose but. So, OCP, by the way, David on his spare time, he's leading the OCP open compute project and the Chairman. So, I really appreciate you break away from the show and come to join the panel. But the standards define some common ground so that the innovation ecosystem and innovating in the most efficient way. But with the accelerated pace of innovation that we talked about early on, you have to engage to allow some purpose boot and so the engagement with hyperscalers, they have the scale.

They move very fast. They know exactly what they need for their architecture, for their networking. So, that's what we found this new way of engagement is a combination of both standards and early engagement. So, that's a part, as I talked about the engagement model. We changed through the Semtech Rising phase is that having the end customer engagement to define what the future generation, for example, right now, the SPO and NPO, the NPO probably still doesn't have a standard yet. The standard pinout and all of this connectivity requirement. But people started innovating already.

There may be some form factor will become the de facto standard through this course, but you cannot sit back and waiting for industry to argue out and define a common standards and then you develop and that will be just too late. So, that's a challenge for us. That's also excitement for us to see that connectivity is finally is not just a conduit to connect something, is an enabling factor for the uptime efficiency of the AI data center.

Yes. So, maybe, Alexis, turning to you again, and you highlighted a little bit on this before, but maybe just kind of give us a level set of where we are today in the journey between pluggables going to NPO to CPO, XPO and so on. So, where are we in that path? And as we get to CPO in particular, what are going to be the key things that you think will determine success for the suppliers because it's going to take a lot to get embedded altogether?

Yes. I mean I think this really comes down. From my perspective, it's, on having operated large-scale infrastructure, it really comes down to an operational metric. What people are trying to optimize for is the uptime and the reliability of the infrastructure itself. So, as we look at the decision points and we look at where the failures are in the data center itself, there's obviously, the entire industry works on pluggable optics today. But we're seeing the switch over to co-packers optics.

I know that there's a big debate on co-packaged optics versus NPO. But when you step back from that, both of them require removing the entire blade to resolve. So, my personal view is the Leap drug to co-package optics because it really comes down to a mean time to repair blast radius mean time to repair and how do you replace that blade. So, I think the ultimate reliability of a co-package optic solution will be higher, and the goal of the industry is to bring that to a standard silicon level reliability and then you still are using these, the optical light source is still a pluggable solution.

So, I think that, that light source remains field repairable. And then it just comes down to the architectural decision of how many light sources per pluggable unit. So, not to make it too technical. But I really, when you step back, it's, I do see the evolution from pluggable to co-package because it's a significant improvement in and it's at least a 30% improvement in power consumption and then the signal integrity and the performance is extremely high. Now there is, I think there will be people who choose to go to the NPO route because it's more comfortable. But we'll see every different hyperscale adopt a different solution on that. But ultimately, that would be my own personal view.

Yes. Did you want to add something, Hong?

I agree. I mean, ultimately, the CPO provide the best power efficiency, people do per bid level, the best integratability and also the stability. People have then kind of in a way, misconception, this is serviceability of reliability, what is the laser having the reliability issues. Well, traditionally, laser has been the route of our evos. If the optical transceiver failed 9 out of 10 times is because of lasers, this state is different and laser reliability is not easy.

What Alexis mentioned about the uptime, the reliability, it's more the network, the linked flaps and had to say this when you have the DSP, when you do not test out the corner conditions in the network, it's prone to link fabs. So, you don't bring the optics so close to the xPU, you don't have, you don't need a DSP, you don't have the chance to kind of getting that complication integrated. So, I think that's great. NPO in the meantime, is a interim open ecosystem play.

So, on CPO, they are establishing the entire manufacturing ecosystem because based on the silicon photonics, it is not as mature as a digital logic from design, from packaging ecosystem. But the industry is building that and they're catching up so quickly. I have no doubt that CPO is going to happen in the future. NPO In the meantime, provide a viable interim solutions allow different hyperscalers to get their customized xPU interconnected. And while they don't have an industry standard again defining pin-out and mechanical key power cells.

Yes, agreed. And it feels to me like next 5 years, CPO is likely to be 1 of the most disruptive things we see here in infrastructure. Maybe Alan, from your point of view, I recognize it's sort of changing, but where do you think the knee of the curve is for CPO deployment? Like when does initial ramp really happen? And when is it just catch fire?

Yes. The initial ramp is now, right? There is CPO in production. And then I'd say, decent or significant volumes beyond proof of concept. The inflection is going to be second half of '28 or into '29 where that inflect. There's kind of 2 bottlenecks there. The first is manufacturing CPO is new, and it takes a lot of new tools in order to get that through manufacturing test qualification. That's new. The industry needs to grow into that. And then secondarily, you've got a time into these road maps, whether that's on the, we'll call it, network ASIC or the xPU GPU, right? And so that kind of creates a question like, do you go NPO to CPO, do you jump right to CPO? That's really going to be a personal decision of each hyperscaler or each accelerator. But the inflection second half of 28%. And as you mentioned, like it's going to be highly disruptive, but in a massive opportunity way.

Yes. So David, congrats to you on OCP given your role. It seems like it's been just a home run and you've outgrown this area. So I understand it's moving in the future. But it seems like 1 of the key things coming out of the show has been Micro-VCSELs versus other approaches. So do you have a view on that, Micros VCSEL MicroLED? How do you think that ultimately plays out and recognize that it's still early, but decisions are needing to be made now for these ramps?

Decisions are being made now. And if I kind of put my OCP hat on, generally, our approach is to allow the ecosystem to debate and arrive at a solution that there is a kind of greater alignment on there, it's probably not the way to go. It's probably not the fastest way to get to what ends up being a standard. But we find that it's the most enduring way because you've allowed the technical debate take place, and you've allowed the voices of support and opposition to kind of bear out. So, and OCP is obviously really well suited to allow for that to happen as various work streams of companies come together, collaborate and ultimately turn that into actual products that enter the market.

So, very nice consensus building thought process. Alexis is a deep technologist, or Hong, maybe the 2 of you, which you when you think of it from just a pure technical perspective of solving all the challenges that would exist, which do you think is going to be, which I think wins out in the end, recognizing we'll coalesce people's perspectives, but the technologists new, where would you think this shakes out?

Between?

Micro-VCSELs, Micro-LEDs as an example.

Micro-VCSEL's, Micro-LEDs.

Slow and wide in for scale up. So, now they seem to be the recent trend is the argument is to microLEDs, you leverage the consumer electronics industry, the MicroLED for display you do the slow and wide. Do you know I know your opinion on that.

So, yes, so I mean, as a pure technologist as an individual, I believe, in the Micro-VCSEL or VCSEL approach or a Micro-LED approach, but that's just because of the signal integrity and the high-bandwidth communications. So, also, when you think about the challenges of the connector and how large the connectors need to be, it's really about system in, but that's, personally, I also lever towards the higher bandwidth solutions, higher bandwidth, less connections because so many fibers, so much connection I've just personally seen inside the data centers and the racks how difficult it is. But what is your opinion Hong?

Yes. No, Alexis, I totally agree in a slow and wide it needs to be reasonably fast. If it's too slow, you're adding a reverse gear box you're getting the native signal converted back to slow signal in the other side to do the other conversion is inedibly adding more power consumption. Kind of, in a way, fit the purpose. And, but of course, the better to be fast and wide. You may need that. And that is what's really the OCI is promised to deliver this 8 channel, DWDM later on 16 channel, even had the promise to do 32 channels.

Each one go with a 64 gigabit or 128 gigabit per second that provide high bandwidth and high bandwidth density too. But I do see that the industry thinking about all the different ways to innovate not only for performance, for availability, for the cost, for reliability. Those are really the things that OCP and a consortium to try to promote, right, not just focusing on technical aspects, but the totality of the supply and the viable solutions.

Yes. Okay. Maybe a question for both Alan and David here. As we look at the transitions that are happening here from like 400 gig to 800 gig to 1.6T, 3.2T, the cadence is shrinking. And so that's great. We're ramping stuff faster than ever before. Is that sustainable? Or is there something else we have to give maybe from a market perspective and then from a user's perspective, David?

I'd say from a market perspective, it is sustainable, right? 800 gig is over 2x the size of 400 gig in terms of volume, and we can expect that at 1.6, we will be at 2x plus the volume size there. So the market is very much able to absorb those types of volumes, but we are running into a little bit of a dilemma wall there. We've more or less maxed out the radix from a switch perspective, and we've maxed out the beachhead from an ASIC perspective inside the server. So that requires the more intuitive ways. That's why there's so much excitement over micro VCSEL this year, right? It solves some of those beachhead dilemmas.

So the market can absorb it and the volumes are absolutely there with the cadence. The cautionary tale is if you're at 1.6T, how much 400 gig are you going to have? So you ultimately want to make sure that you have a home for those older technologies as we go through these faster adoption curves.

Yes. And from a user perspective?

Yes. I mean so I'm obviously not going to answer the sustainable question. But the way I think about it is that evolution of the network is kind of matching the evolution of compute. And that evolution of compute is largely matching what our kind of needs are. So if I think of

[Audio Gap]

Left. And it's been a great discussion. There's obviously lots of complexity, lots of opportunity here ahead in this whole opportunity set for networking to drive AI infrastructure and all the benefits that it's going to deliver in the future. As I said, it's a complex world. There's lots of the market is concerned. What's the one just assumption you think that the market just has wrong today about -- because to me, it's obvious where things are headed, but the market doesn't always see it the same way. So what's the one thing you think is an assumption people are making that just isn't -- doesn't hold water. Starting with Alan.

Yes. I think it's that these are binary answers, right? We're going to get the CPO. There's no doubt there, and it's going to happen, I'd say, roughly in that 5-year horizon. But the misconception is that these things are binary. Copper goes away, copper is not going away. If we do micro VCSELs, we're not getting rid of transceivers. So these aren't going to be binary events. So the right network for the right part of the server or the right part of the memory, and that ultimately leads to a lot of markets running in parallel.

David?

I'd be -- I'm humble enough not to make a statement about what the market is wrong. I generally observe the market to learn from -- yes, and I, over time, find that it's actually more pressing than most forward-looking statements out there. If I were to add one area where I think it's probably underappreciated by the outsiders is the amount of innovation that's happening in the hardware space right now at all levels. There is more emerging companies, there is more start-ups. Many of them have fantastic ideas they're working on. And the rate of innovation is kind of unmatched to anything I've seen in the past 25 years I've been in this space. And I think that's generally underappreciated by the outsiders, those outside looking in.

So I agree with what both of you have said. I mean I would take it just at a slightly different angle is I'm an incredible optimist for the power of AI for human outcomes. I think this will be an era of incredible abundance that we're about to unlock. So it may not be what the market is getting wrong, but I think that there's a lot of fear in the market that could really be converted to optimism. It's -- we're at the era and at the precipice of an incredible time.

And it's going to require an incredible amount of compute and compute access. So just like -- and it's the opportunity -- I fundamentally believe it's an opportunity for all of humankind globally to seize this moment and really unlock the power of AI. So that is -- that's more of a wish on my part. But I get to operate at the very frontier of AI. I see where it can go. And so I really am very bullish because of that. And I think that we're going to unlock this moment. This moment will be really better appreciated maybe in the coming year.

Hong, you did the last word.

Yes. So I think the disconnect, what I see is today's assumption of sustainable or not is based on the technologies today. But we continue to innovate for example, you're going to be hearing this afternoon, XPO increased density of 4x power efficiency can be improved. So the cost of building data centers, the power consumption, the labor and footprint you need is all going to be revolutionized. People have not factored that into their model and look at the macroeconomic situation.

Another thing is a disconnect between the macro concerns and really, we are in a trench building the product, and we have the great visibility in the daily drill, we have to go through expediting. So I hope sometimes it will be converged somewhere. But I'm very excited about this. I do believe, as I said, the best is yet to come.

Yes, I couldn't agree more, and you can't change the world in a straight line. So -- but I think there's just tremendous opportunity ahead. I want to thank the panel for just great insights. Thank you all for being here as well.

Thank you...

Please join me in thanking the panel one more time for a very engaging discussion. So we will take a short break, about 10 minutes. We'll resume at 11:00 for those of you on the webcast, 11:00 Eastern. Imran Sherazi will go into the data center strategy at that point. Thank you.

[Break]

Please welcome to the stage, Senior Vice President and General Manager of Signal Integrity Products, Imran Sherazi.

Good morning, ladies and gentlemen, and welcome to the Data Center Deep Dive. My name is Imran Sherazi, and I'm the General Manager of the Signal Integrity Products Group. And I'm really excited to be up here this morning to share with you our vision for Signal Integrity in the data center and why we believe passionately that the market is moving towards us. I've had the pleasure of meeting some of you before, but I see lots of new faces. So welcome, everybody. Thank you for attending, and I look forward to meeting with you afterwards, perhaps over lunch or maybe at the demo stations at the back of the room here.

I'd like to begin today by sharing with you our incredible history in Signal Integrity. Many of our team worked together in the late 1990s in the telecom era, developing TIAs at 50 megabits for OC1, which was the first optical standard. The business in its current form came together in 2002. And the first success was in the telecom market at 2.5 gig and then at 10 gig, but we quickly pivoted to 10-gig Ethernet. And I still remember when we hit $10 million in revenue.

We were a small group back then and the lead designers, some of whom are still with us, were so thrilled that their designs were competing in the open market, winning sockets, and being installed in networks that were becoming the Internet. We used to tell our friends and our families that their e-mails were sent and received over our chips. We went on from 10-gig Ethernet to address fiber-to-the-home, then 4G wireless, and then the cloud data center at 4x 25 gig.

We developed all these product generations. But it was the passion for winning for seeing their work become part of the network that really drove those designers to the innovation and success that we've built the business on. As the line rates went up, we became the analog Signal Integrity pioneers. And I don't say that lightly. Signal Integrity is a dark art. That's why it's called a dark art. You won't find it in textbooks.

Groups like ours were developing the art through generation after generation of product, constantly pushing the boundaries and learning hard-won lessons. As we grew, new grads came in, they were coached, trained, taught that dark art by the senior engineers and then they, too, became part of that cult of signal integrity. And here we are now, delivering 200 gig at scale, introducing 300 gig, and designing 400 gig.

When you look at this graph, you'll soon realize that the AI data center is literally what our business was built for. And that's why we passionately believe that the market is moving towards us. We talk a lot about Signal Integrity. I'll talk a lot about it here. So I thought it would be a good idea to give you an introduction to some of the principles of Signal Integrity. What is signal integrity. It's about impairments, the set of impairments that impact a signal, especially as that signal propagates across a channel and how we deal with those impairments. So let's have a look at one of those impairments, attenuation.

How do we think of attenuation? I'm speaking up here, and you can hear my voice quite clearly. But if I start to move away from you, my voice will get quieter and quieter and quieter. That's attenuation. From my position here to your ears is the channel, and the signal is my voice. And the further I am away from you, the more my voice will be impaired. And that's exactly what happens to a digital signal in an AI data center today. It suffers attenuation the further it propagates across the channel.

And what you see in the diagram is a healthy digital signal to start with, a transfer function where there's some attenuation and a very diminished signal at the end there, so low that it doesn't cross the decision threshold. And so it doesn't register as a signal at all, just a series of zeros. That's attenuation. How do we deal with it? With gain, like a microphone. We add gain to the signal bring the signal amplitude back up so it can exceed the decision threshold and those 0s and 1s can be read correctly. Let's look at a second impairment, bandwidth limitation.

You're all quite familiar with bandwidth limitation. Every time you're on hold on a call with a call center of some kind, you listen to music, and it doesn't really sound like music because that channel, that telephone line channel, is very heavily bandwidth limited. The human air can hear from 20 hertz to 20 kilohertz. But that telephone channel is limited from 300 hertz to 3,400 hertz, just a narrow slice of the audio spectrum, a narrow slice of that high-fidelity music spectrum. So it doesn't even sound like music. It sounds flat. The lows aren't there. The highs are heavily attenuated.

It doesn't really sound like music. It's almost as if the words are smearing into each other. And that's what you see here with a digital transmission. The effect of bandwidth limitation on that digital bit that passes through the channel is for the rising edge and the falling edges to spread out because the high-frequency content has been attenuated, bandwidth limitation. It's the high frequencies that don't make it through. And so you can't get that nice square wave.

Instead, you get this spread out pulse, and it's reduced in amplitude because the high-frequency content has gone. And so some of that energy is lost. And as you can see, what started off as a 0001 isn't really being picked up. That force 1 isn't being recognized by the receiver due to that bandwidth limitation. And what do we do to overcome bandwidth limitation?

We use linear equalization, apply EQ to the signal in order to prop up that curve that you see there to get more gain at those higher frequencies in order to allow those higher frequencies to pass through and maintain the alacrity of the signal. Let's look at a third impairment, non-linearity. This isn't a channel impairment. This is an impairment that occurs in active devices, devices that don't have sufficient linearity. What's a good example of that? You take a guitar, an electric guitar, and an amplifier.

As you turn up the volume on the electric guitar, the volume from the amplifier increases until you overload that input stage to the amp. And that's when you get distortion. And of course, guitarists love to whack up that volume, get that lovely distortion. We all enjoy it, but not in Signal Integrity. Here, we don't want that non-linearity. We want our devices to operate in the linear region. And how do we do that? How do we ensure that? We use automatic gain control.

So we control that gain. We make sure we've got enough gain for small signals, but we reduce that gain when the signals are large. We can also use pre-compensation. Finally, let's look at noise. Noise occurs in every system, and AI data centers are not an exception to that rule. You get noise from the power supplies, you get noise from adjacent racks, you get noise from the fans in the back of the power supplies, and you get noise from adjacent lanes. Every channel, every lane that's transmitting a high-speed signal, is radiating RF noise.

And the channel that we are sending our signal down is susceptible to that RF noise. And so what you see is noise on the 1s level, noise on the 0s level. How do we think about noise? Again, I'll go back to my example of my voice. You can hear me quite clearly right now. But if somebody were to start operating a jackhammer by the side of the stage here, I might get drowned out. That noise level, if it's greater than my voice, if the noise is greater than the signal, that's a low signal-to-noise ratio. My voice is lower than the noise.

Ideally, that noise wouldn't be there. You can hear the fans humming in the back there from that 64-port switch. If that were the only noise, my voice can overcome that noise, a high signal-to-noise ratio. And so every signal is susceptible to that noise. If that noise is very significant, you might see that noise on the 1s level or that noise on the 0s level being so significant that it exceeds the decision threshold. And that would result in a erroneous bit. What's an erroneous bit? We talk about BERs, 1E-9, 1E-12.

An erroneous bit is a BER. 1E-9 is 1 in 1 billion bits, 1 erroneous bit out of 1 billion bits transmitted. 1E-12, 1 erroneous bit out of 1 trillion bits transmitted. How do we deal with noise? We use low-noise design for a start. Secondly, if the noise isn't in band, if it's not at the same frequency as the signal, we can filter that out. That's easy. It's the noise that is in band, that's on that signal, that is the real challenge when it comes to managing Signal Integrity.

I hope that's been helpful for you, a quick overview. It's obviously a lot more complex than that. But let's see how some of these principles apply in a real-world data center application. So here, we're showing a GPU ASIC, talking to a switch ASIC at the far end of the screen. That, I'm sure you can imagine today is a very common use case in a data center. The GPU ASIC will output a nice clean waveform as you can see there. Maybe some noise gets added.

And so that Signal Integrity gauge that you can see below the signal starts off firmly in the green. It's a nice clean signal. Some noise gets added. That signal integrity gauge moves to the right, but it's still in the green. Bandwidth limitations of the channel, as we discussed. That signal integrity gauge is now moving into the orange zone. This is starting to get challenging for the switch ASIC at the far end to be able to resolve with a good bit error rate. Finally, attenuation. Now that Signal Integrity gauge has moved right into the red.

And if you compare that signal trace in the red box with the original transmitted signal, they barely resemble each other. And so the switch ASIC at the far end, there's more distance to go yet, is going to have a real challenge in resolving that signal with a good bit error rate. So we introduced a linear redriver, the Semtech IC, one of our copper edge ICs. That linear redriver works its signal integrity magic, some of the approaches we discussed on the last slide, plus many more.

And you get a nice, restored output signal. It's not perfect, but it's restored and it can move -- it can propagate some distance down the channel, be received by that switch ASIC, get resolved without bit errors. Let's look at another example. We've all heard about retimers. Here, I'm showing at the top, a retimer example and at the bottom, a linear redriver example. What's the difference? What does the retimer do in that chain? You may have already spotted it, but I'll highlight it here.

The retimer has effectively reset all the impairments before it, all the impairments of the channel, it's recreated that signal. If you look at the technology inside a retimer, inside a DSP, it's essentially the SerDes technology that's used in the ASIC, where the signal originally originated from. So it can completely reset that signal, recreate it perfectly, and send it down the channel. How do we think about that, retimers versus linear redrivers? Picture an example of a flight from Sydney to London.

The retimer example would involve a stopover in Singapore. The flight would land, that's latency, would have to take off again, consuming a lot of fuel, that's power, airport charges, that's cost, the footprint versus the linear redriver example is like a lighter aircraft taking off from Sydney, flying direct to London. How? It gets a boost along the way, a strong tailwind. And if you think about that boost, it gives you the opportunity to maybe take off a little lighter.

In other words, an opportunity for the host ASIC to perhaps not do as much heavy lifting as might have been required in the past. And that's the beauty of linear Signal Integrity and why linear solutions are becoming more prevalent in the data center today. The way we think of it is like this, linear when you can, and retime when you must. Thank you, Jensen. Let's look at Signal Integrity by function and understand some of our key products. Linear redrivers, we just talked about those. We saw how they work. And essentially, they extend the link budget of an ASIC. That's the value proposition.

At their core, they do one job to invert or undo the effects of the channel. We just talked about those various effects of the channel. That's what a linear redriver does, does it very efficiently, very elegant solution. If we look at the function of a TIA, a TIA takes the signal from an optical fiber that's been converted by a photodiode into a tiny electrical current on the order of microamps. And that current has to be converted to a voltage, and that's transimpedance, the conversion of that current to a voltage. That's what a transimpedance amplifier does.

It undertakes that conversion while maintaining very low noise, very high bandwidth -- pardon me, not very high bandwidth actually. Some of my colleagues would berate me for that comment because very high bandwidth is actually what you don't want. You want high enough bandwidth and high gain.

Finally, optical drivers. These are voltage drivers that drive lasers or modulators, preparing that electrical signal so it can be converted to light through a laser on through a modulator. That's a function, the laser driver function that we've seen emerge from being integrated within the retimer to now a stand-alone function because line rates have increased and the signal integrity challenges have become too significant. It's very difficult to use an integrated driver now and increasingly so as line rates increase.

In all of this, signal integrity is like a foundational requirement upon which we really have to get creative. My design colleagues have to get really creative. Ever-evolving architectures that perform each of these 3 functions and push the limits with every generation. Use copper when you can, use optics when you need to. Our view of the data center is completely aligned with that. This is how we think about the portfolio, scale in or scale up versus line rate.

The boundary that you see here between copper and optical may shift, especially as line rates increase. And that's okay by us because for us, optical or copper, it's all Signal Integrity. What we're really passionate about is having the best solutions, whether it's linear redrivers for copper, or TIAs for optical, or lasers and photodiodes for anything optical from scale up to scale out. That's what we love to do. That's all Signal Integrity to us, and that's our mission.

Let's take a look at our CopperEdge product line and why we believe we have the right to win. The industry didn't believe in redrivers. We partnered with a major AI supplier and proved that technology at 224 gigabits. We proved that we could add 20 dBs of link budget to an ASIC at 224 gig. And we're shipping that solution in volume today. But what we're really excited about is our 336 gig redriver. And here, we're very proud to share a recent proof point.

We just learned that our part is the only solution out of 6 competitive redrivers that currently performs in an AI data center application. How did we get there? 336 is an emerging line rate. It's bleeding edge. So our engineers worked very closely with our foundry partner on the technology and with the customers' engineers to understand the high-speed system trade-offs and how we could push the limits of the process technology in order to get the best possible performance at 336 gig. It's easier said than done.

It took lots of high-speed simulations, the requisite whiteboarding of concepts. But of course, you can imagine that's what you always strive for is getting that dynamic, getting engineers in a room together around a whiteboard. Whenever that happens, we know we're going to end up pushing the limits of the technology. We're going to agree the right trade-offs, where we need extra performance, where we can pull from, where we can push to and ultimately deliver this kind of success like 336 gig redriver.

We're now working on other opportunities using the same approach, same hands-on approach for line card applications, active backplanes and ACCs. And the market is confirming the thesis. As line rates increase, AI data centers need linear Signal Integrity. Our FiberEdge product line, TIAs, and drivers. As I mentioned at the start, we have decades of experience in designing TIAs from our first 2.5 gig products to 20 years later, 200 gig. But what does that really mean for us? I want to draw your attention to the first box here, #1 in 100 gig per lane TIAs.

That wasn't our first 100-gig part, but it was our best. And our customers have confirmed time and again that it's the best part in the industry. And that's actually not new for us. We did the same at 2.5 gig. We did the same at 10 gig. We did the same at 4x 25 gig, and we're going to do the same at 200 gig. And for us to take our market share from 18% to 15%, the only incentive we had to use was the performance of the part. At 200 gig, our shipments have only just begun, but our engineers are super excited about what they're developing.

We're super excited about what they're developing, and they're really eager to surpass that 100-gig record. And that 200-gig solution, together with our 200-gig laser drivers is going to be the foundation going forward, initially in 1.6T modules, but then quickly in other form factors, NPO, XPO, and other highly integrated form factors. Of course, we're looking forward to that challenge and the ensuing success. Let's talk about Semtech Photonics, our latest offering.

As you can imagine, as TIA and driver designers, we've spent decades designing the parts to work with the available optics. Now we actually get to work with the optics designers in-house. We can both optimize our designs. We can deliver the highest performance, the most efficient solution end-to-end. Three reasons why our Photonics portfolio wins. Number one, power efficiency, 44% power conversion efficiency at 25 degrees, less than 0.8 dB of coupling loss versus about 1.2 to 1.5 that's typical in the industry today.

That means we save 40 to 50 milliwatts of power in an 800-gig module. Number two, signal quality. The lasers have very low RIN, very tight line width. They're essentially coherent light-ready today. And number three, we're a U.S. supplier of indium phosphide. With HieFo, we have 46 years of photonics chip experience. We're vertically integrated from epi to chip on submount. And with DTC, we're broadening the portfolios to include photodiodes, too. And of course, all of this fits perfectly in our scale-up, scale-out, scale-across portfolio.

Delivering performance and quality at scale. I'm sure you'll all agree, that's the order of the day. We participated in many very significant build-outs throughout that 24-year history of ours. The initial hyperscaler deployments in the data centers in the 2010s, the ultra-high-volume fiber-to-the-home build-out, the very high-volume build-out of 4G cellular. We've been delivering high performance at scale for 20-plus years. It's part and parcel of our long-standing relationship with our cloud partner, Tower. You've all seen the press release earlier this week. That was just the icing on the cake.

Our supply agreement with Tower supports our financial plan all the way out to beyond FY '30. We were actually Tower's first optical customer for their silicon germanium process all the way back in 2002, and we've been a lead customer for Tower since then. Our engineers have been collaborating ever since. They've been pushing silicon germanium to the next level of performance for 7 generations of process technology, driving up bandwidth and driving down power all the way from 20 picojoules per route hertz to 1 picojoule per route hertz and even lower.

We began at 2.5 gig, and now we're deeply engaged in pushing to 400 gig. So that's the underlying foundation of the technology and the innovation. But how do we leverage that for our customers? Whether we're working with an optical module designer or a hyperscaler engineer, we're always striving for that example that I gave you earlier, that same approach that we applied at 336 gig in the linear redriver development, getting the engineers around the table, getting them around a whiteboard. You can't push the limits of technology by throwing a solution over the wall.

It doesn't work like that. Whenever we work together at the engineering level, that's when we know that we can really change the world. Let's look at the various form factors that are emerging for 1.6T and beyond. We call ourselves form factor agnostic. Whether it's copper, FRO, LPO, LRO, employing our linear redrivers or our TIAs or our drivers, these are all opportunities for us. Let's look at highly integrated or look at the evolution of integration, should I say, from LPO through NPO to CPO, the module sits on the PCB, is plugged into the PCB and the ASIC sits at some distance. In the NPO application now, you have the addition of an optical engine that sits on top of an EIC, an electronics IC that would contain our laser drivers, our TIAs and perhaps even our linear redrivers. And ultimately, the objective there is to increase the density of that module. So today, you're all very familiar with a 1.6T module, an OSFP module and NPO would nominally instantiate 2 of those, but even up to 4 of those. And CPO would, of course, go even beyond that.

Let's look at an active copper cable. This is one end of an active copper cable. That copper cable has multiple strands that you can see here are terminated in the PCB and that module can plug to a rack in a data center. Active backplane. I mentioned this briefly. Here, you can see our linear redriver chips sat on the backplane itself in order to provide that tailwind, that boost that I talked about.

A traditional optical module as is being shipped today, OSFP 1.6T, employing our various components. An example of an NPO, 3.2T, you can see the scale. It's probably about that big. It's very compact, fully linear, low power, high density in order to work its way into scale-up in the data center. And then XPO, which is a very high-density module form factor, but I'm not going to explain it to you. We talk about partnering with our customers. We have here today Sunil Priyadarshi from Arista, Senior Director, the force behind XPO. So I'd like you all to welcome Sunil to the stage. He'll join us, and he'll talk about XPO. Sunil?

Thank you, everybody. I'm very thankful to Hong Hou for inviting me. I won't talk a lot about the -- why we need the higher bandwidth for the AI. Panel was very clear about that. Actually, I'm holding this is, most of you are aware is 1.6T. This is the most successful MSA in the optics industry. Even this year, there will be 100-plus million of the optics, 800G, 1.6T will be made. Next year, that quantity could go to 200 million. So there is a vibrant ecosystem. There are TIA driver, lasers, PDs, DSPs, everything. And there are 100-plus vendors that make these optics.

So what is the problem? Why we cannot keep using for the AI? What is the problem? So problem is very simple, density, bandwidth, latency and power. Let me give you the example. If you want to connect the 1024 GPU, if you assume each GPU has the 25T of bandwidth needed, if you're using these optics, you will need the 16 racks. So now you can think about you have the 4 racks of the GPU. To connect them, you need the 16 optics rack. So that makes your footprint pretty big. Your interconnect length will be longer, and that give you the latency, higher power. Probably copper won't work. You need the more expensive interconnect.

So how we deal with this thing, okay? So -- but if you can get something for the AI application, which is very close to this because this is pluggable, very good ecosystem, easy to build, a lot of -- actually, I have seen so many videos, automated factories, which are built now state-of-the-art. So why don't we get the optics, which is more close to this, not like innovation could be very painful, but time to the market is what we need because the 200T switch, which is just coming next year. So we need the optics, which can meet, give the highest density if you go with the 200T.

So that's what is the XPO, okay? It's equivalent of the 8 of this. So bandwidth is 12.8T. So now let's do the math again. So if you use this XPO for the -- again, for the same like 1024 GPU to connect, you just need 4 racks. So the reduction in the footprint is like 16 versus the 4 and so XPO can give you the 200T, you can get into 1 OU. If you want to have the 200T switch with the OSFP, you need the 4 OU. This gives you the 1 OU. If you look at the rack level, this will give you a full rack will give 1.6 petabits, this will be 6.5. And that's what you need for the AI.

So what are the other things for the AI? It gives you the bandwidth, it gives you the density. Let's talk about the power, okay? So XPO actually, we have proven already it's with a full retime to half retime. And we also recently presented some data. If you use XPO in the switch, you can do the linear. So XPO does meet the linear on the top. XPO use the same ecosystem. That is the reason that XPO MSA, we got 175 vendors joining the MSA. This is the highest any MSA like the company joined a single MSA.

Why? Because people are already building this one, they know how to build it. XPO also, they can build in the same time. They don't need a lot of innovation time. That is already existing technology. It made the density same like the CPO. It gives you the -- like the wider ecosystem than CPO, you are just tied with a few like the very small ecosystem. That is the reason the XPO is getting a lot of momentum, and it needs all the optics for the AI now. And like XPO can also go to 400G, then this form factor can give you 25T.

So now think about the -- why optics bandwidth is needed so large because now you think about the 100 -- 1,000 GPU to 100,000 GPU. So -- and in the future, like the GPU bandwidth is going to increase. Now it is 25T. Future, it could be 100T. So the bandwidth needed is not linear to the number of GPU, like if you're increasing 1,000 GPU to 100,000 GPU, you're increasing like 100x, but future GPU will be also needing more bandwidth, like 100T -- 25T to 100T. So this means bandwidth -- network bandwidth has to go 400x because math is simple, very simple.

I think a lot of people talk about CPO. CPO ultimately will happen, most probably at 400G. And actually, in the meantime, if you can use like best of the pluggable and best of the CPC because XPO is like the NPO because the main purpose is how you can make the insertion loss lower. And XPO use the CPC and the flyover cable. So as you know, the flyover cable can have 4 to 5x less insertion loss than the regular PCB. So this is kind of sweet point for the current AI need, the optics bandwidth right away, so -- and easy to build. And next year, you will see we'll see -- you'll have the 200T switch in OU using the XPO.

And one of the more things is that CPO, there are a lot of new things to be learned, especially putting the fiber inside the box. That is the most delicate part of optics. Everybody knows that. But XPO can avoid everything. It has a coaxial cable inside. It gives you the lower insertion loss. It supports the linear channel and future XPO can support 25T, then it will give you even 400T in 1 OU.

Very high density.

Thank you. Thank you very much.

Thank you, Sunil. Maybe I can take this one from you.

Yes. Yes. Thank you.

Awesome. Thanks so much. Please give Sunil another round of applause. It's hard to believe that, that XPO module instantiates 8 of these 1.6T modules in a pluggable form factor. That gives you density, it gives you pluggability. It gives you the ability to switch in and switch out if something were to go wrong. Very exciting. And again, for us, it doesn't matter what the form factor is. For us, the challenge remains the same, and we're extremely excited about it.

Let's look at a form factor like this and how it fits in an AI data center. So what you're looking at there on the left-hand side is an optical module, I'll put it the right way around, like this. And what you can see inside that optical module are our products. So 2 TIAs, 4 channels each at the top there with photodiodes, 8 of them, a driver IC, a laser driver IC, 8 channels and a CW laser source. That's the optical module. The active copper cable looks very similar. I would encourage you all to participate in the show and tell at the back of the room after lunch, you'll get to see active copper cables and optical modules like this.

But inside an active copper cable, as I mentioned earlier, the various channels are implemented as single strands of copper terminated on the PCB and then our linear redriver chips, our CopperEdge chips sit on that PCB. But where do these go? How do these fit inside a data center? Let's have a look at that. These repeat across the data center. So there you see the optical module and our ICs fit into that, as I just described.

But just as an illustration, 20 of those modules would fit inside a compute tray about 16 inches wide. Those compute trays would sit in a rack, 8 of them in 1 rack, a little bit taller than myself. So 20 modules in a tray, 8 of those trays in a rack and 64 of those racks in a cluster. It's like a corridor of racks actually in the data center that are on both sides, you're surrounded. And there are 49 of those clusters, at least in this illustrative example that we're providing, 49 of those clusters in a data center.

So we start with 2 of our TIAs, a laser driver, CW source, photodiodes, 20 of them in a module -- 20 of those modules in a compute tray, 8 of those trays in a rack, 64 of those racks down a row and 49 of those rows effectively. And the net result is 500,000 modules in this example of a data center. It's a staggering opportunity. The whole point of the data center is to move data, but it's signal integrity that enables the movement of that data. And as I've said a few times, especially as the line rates go up.

It's a generational opportunity for us. That's how we view it. It's everything we've been doing to date culminates in this. So I'm confident that by now, you've appreciated that we have both the experience and the expertise to deliver success. But what does that mean for you in terms of dollars? We expect to more than double our data center revenue this year and more than double again next year. That's what we call a generational opportunity. And that's a generational opportunity that we've been able to realize as a team, and it's a very fitting milestone for such a passionate team that's been winning together for 24 years. And honestly, it feels like we're just getting started. And I really encourage you to meet some of the team members at the back of the room here to feel that passion and energy that they have for achieving these targets.

Everything I've just talked about from CopperEdge retime when you must and redrive when you can to the design team's excitement around what we're building for 200 gig to the photonics opportunities that are just getting underway and our content opportunity in the optical module growing up to 10x. We're literally having the ride of our lives, and this is giving us great line of sight to our FY '30 financial targets. We expect to outgrow the SAM, all these new opportunities that we're engaged with, all these solutions that we're making happen, extending the reach of copper, participating in all the different module form factors, developing the best solutions for all of those form factors from a common base of -- to start with 200-gig TIAs and drivers, getting the right footprint, getting the right pitch to proliferate into all those different module form factors. And all of that's giving us great line of sight, allowing us to outgrow the SAM ultimately to get to $2.1 billion by FY '30.

So some key takeaways. Signal integrity is what we've been doing for decades. And the AI data center opportunity, it's not the first huge market inflection we've succeeded in, but it's definitely the biggest. And what's even more exciting this time is the opportunity is not just optical. It's both copper and optical, and we're just as passionate about driving both of those solutions for both of those. And again, the whole point of the data center is to move data. And it's signal integrity that enables that movement of data at ever-increasing speeds. It's a generational opportunity. And so watch this space, enjoy the ride.

Thank you, and I look forward to meeting many of you over lunch and at the demo stations after lunch. Thank you.

Thank you, Imran. At this point in the program, we will take a break for lunch. Lunch is served across the hall. I would encourage you to echo Imran. There's a great opportunity to see the products in action. So during the break, use the opportunity to do that. We will reconvene for those of you on the web in the room at 1:00 p.m. Pacific. Thank you.

[Break]

Good afternoon, everyone. Just a reminder, demo hosts will scan the QR code on your badge at each station that you visit. Once you've collected all 6 scans, stop by the registration desk to claim an exclusive gift as our way of thanking you for your time. Thank you.

[Break]

Please welcome to the stage Semtech's Executive Vice President and Chief Financial Officer, Mark Lin.

Good afternoon, everybody. And just allow me to convey how truly grateful I am for your attendance, both in person and online. I hope you come away today feeling that this was time well spent. For me, seeing Imran share his passion for signal integrity with all of you is truly gratifying. Imran has done plenty of teach-ins for me. His team has done plenty of teach-ins for me. Hong said, winning is infectious. And to me, signal integrity is infectious. I love this stuff.

I also feel privileged to be one of the voices today, representing my Semtech colleagues to highlight the results of our accumulated efforts, the financial transformation we've been able to achieve together and our path for accelerating growth. You know about 10 quarters ago, we were incurring staggering cash interest costs, which resulted in cash flows that were flowing the wrong way. Today, that picture has changed. We're in a position to raise our Q3 '27 outlook and provide financial targets, extending to FY '30 that shows Semtech's growth trajectory is closer to the beginning and far from the end. So let's get started.

My key messages first touch upon a bit of history aligned to Semtech Rising and moves to our current state of Semtech Transforming and our future state of Semtech Excelling. First, Semtech's financial transformation paved the path for accelerating growth. I believe we first set the company on firm financial footing, and we're primed to capitalize on market tailwinds and accelerate our growth. Second, I believe we've demonstrated consistent execution in meeting our financial targets, a critical track record as we provide targets through fiscal year 2030.

Third, we expect a positive step change in margin profile, supported by growth in data center and LoRa, 2 very high gross margin products. Fourth, our cash flow generation, portfolio transformation and disciplined capital allocation enables organic and strategic investment to strengthen our portfolio value. Fifth, our differentiated portfolio of high-performance solutions paired with strong margins and balance sheet stability is expected to compound into revenue, earnings and cash flow growth, resulting in increased shareholder value. We'll highlight this point through our FY '30 financial targets.

One of the hallmarks of Semtech Rising was to fix the foundation, enhance our capital structure, enhance employee engagement, enhance customer engagement. What I've concluded is that these 3 items are remarkably linked, and I believe we have meaningfully improved the state of all 3. We've derisked the balance sheet through repayment of high-cost term loans. We've managed debt maturities. We've lowered our cost of capital, and we've improved liquidity. About 10 quarters ago in FY '24, the year I joined Semtech, net cash interest cost this company $82 million, and free cash flow was a negative $123 million, clearly unsustainable. Today, net cash interest is a contributor to free cash flow and annual free cash flow as of Q2 of '27 was $193 million.

Likewise, net debt at the end of FY '24 was $1.3 billion with a net leverage ratio of 10.5x, also clearly unsustainable. But at the end of last quarter, those figures are down to $299 million and 1.1x. Overleverage is in our history, but it is history.

Within Semtech's ecosystem, including employees, suppliers and customers, our financial transformation has allowed us to amplify our signal and reduce the noise. First, in R&D, we've been able to accelerate the pace of investment with a laser focus in our high-growth areas and generate top line growth and bottom line return and certainly better return than we are earning from interest expense.

I've been able to witness firsthand this multiplier effect on our engineering teams from our increased but very focused and disciplined investment. After ECOC last month, I made a stop at Semtech's location in Switzerland to visit the LoRa development team. Our increased investment has energized this team, and it's not for the spend. It is because this team understands where the market is heading. They're listening to the customer, and now they're able to accelerate the development of the next generation of LoRa. Our next-generation LoRa device, it really focuses on customer-requested enhancements, substantially lower standby current, which dramatically extends battery life, increasing data transmission rates and all with SDKs facilitating device deployment.

Second, we've been able to contribute to growing EPS as interest savings have resulted in increased earnings. Third, we've been able to substantially increase assurance to our suppliers and our customers, which I believe is critical in today's accelerated product development time lines. Suppliers have increased their credit limits. On the customer side, I have not had a discussion with the supply assurance team to discuss Semtech's financial condition in some time. That's all been taken care of. But more importantly, and I believe most importantly, during customer meetings, our sales and field engineering teams can start a meeting with the customer, those whiteboarding sessions that Imran was talking about. We can start those meetings discussing and leveraging Semtech's portfolio to assist the customer. Those teams don't have to go into these customer meetings first having to discuss Semtech's leverage.

We've demonstrated consistent and predictable execution. Over the last 10 quarters, we've met or exceeded our midpoint revenue and EPS outlook, all while growing revenue in each of those quarters. We met these quarterly targets while growing revenue 66%; adjusted EPS, 1,083%; adjusted gross margin, 470 basis points; and adjusted operating margin, 1,220 basis points. I highlight this track record of execution as I believe it is foundational in giving you confidence in our ability to execute on our multiyear model.

Speaking of execution, we're updating our outlook for Q3 of '27 to the upside. Allow me just to review some midpoint figures for these updated expectations. Revenue at $417.5 million, which would be 56% year-over-year revenue growth. Consolidated adjusted gross margin at 59% and adjusted gross margin, excluding the held-for-sale cellular modules business, at 64%. Consolidated adjusted operating margin at a record 33%, adjusted diluted earnings per share at a record $1.14, up 138% year-over-year.

On the divestiture of the cellular modules business, we continue to expect that the transaction will receive regulatory clearances and that we will be able to close this transaction in our fourth quarter. On portfolio optimization and moving to a higher margin mix. Growth in our portfolio has shifted dramatically to data center and LoRa. In FY '24, data center and LoRa each comprised about 7% of total revenue.

Moving to our expectations for Q3 of '27, excluding the cellular module divestiture, Data center is expected to climb to 40% of total revenue. LoRa revenue is expected to constitute 19% of total revenue. Translating these percentages into dollars. For all of FY '24, data center revenue was $61 million and is now expected to grow to $145 million over a single quarter, Q3 of '27. LoRa revenue was also $61 million for all of FY '24 and is expected to be $68 million for Q3 of '27. Both portfolios are at record levels and contribute amongst our highest gross margins, and growth expectations through FY '30 for data center and LoRa are the basis for our belief that we can grow beyond this quarter's expected 64% adjusted gross margin, excluding the cellular modules business.

Semtech's margin performance is highlighted here. I believe it reflects a clear inflection. I'll cover historical growth and compare to our midpoint outlook for Q3. Q2 of '25 to Q2 of '27, over 8 quarters, adjusted gross margin increased 410 basis points. For Q3 of '27, over 1 quarter, we expect a 950 basis point increase in adjusted gross margin, excluding the cellular modules business. For Q2 of '25 to Q2 of '27, over 8 quarters, adjusted operating margin increased 1,020 basis points. For Q3 of '27, over 1 quarter, we expect an 860 basis point increase. This is translating to adjusted earnings per share. Q2 of '25 had the headwind of interest expense, pretty substantial interest expense, and we reported adjusted EPS of $0.11. For Q2 of '27, adjusted EPS was $0.71 or growth at 9.3x sales growth compared to Q2 of '25. For Q3 of '27, adjusted EPS is expected at $1.14 with growth at 2.7x sales growth over a single quarter.

We're returning to the gross margins that this company has earned before. But this time, with top line growth several times higher and operating margins and earnings per share above what Semtech has previously been able to deliver. Scale up, scale out, scale across copper, optical, photonics. They've all been contributing to Semtech's scale at the revenue level, providing operating margin, operating leverage and earnings growth.

I'd now like to highlight capital allocation as it relates to operating expense spend. The first point is absolute. We focus investment in R&D and allocate resources to high-return projects where we have customer alignment. I think this is evident in the 27% increase in R&D dollar spend between FY '25 and the last 12 months ended Q2 of '27. Our R&D opportunities are vast, but we focus on customer alignment. Our ROI calculations factor where customers have invested resources on the project we are supporting. We believe these factors result in greater assurance in the quality of our ROI expectations.

Even with these opportunities, though, we do maintain an envelope of affordability and have roughly maintained R&D spend as a percentage of revenue over the periods we're presenting here. One way to afford R&D is to watch the efficiency of SG&A and especially G&A. While R&D investment increased 27% on this chart, SG&A increased 8% or 30% of that rate over the same period. We are investing in our commercial team. Listening to our customer is important. But you can see as a percentage of revenue, SG&A declined from 18% to 15%.

Scale is another factor in affordability. Even with increased R&D investment, our Q3 '27 outlook pegs the midpoint adjusted operating expense as a percentage of revenue at 26%. That will be down 910 basis points from FY '25, down 670 basis points from FY '26 and down 600 basis points from the trailing 12 months ended Q2 of '27.

Turning to the balance sheet. We ended Q2 of '27 with $503 million of debt against $204 million of cash and cash equivalents for net debt of $299 million. Combined with $357 million of availability under line of credit, that gives us total available liquidity of $561 million. Our debt maturity schedule is clean, $100.5 million comes due in FY '28 with no scheduled maturities in FY '29 or FY '30, and the remaining $402.5 million isn't due until FY '31. With expected Q3 trailing 12-month adjusted EBITDA of roughly $360 million and growing cash generation, our net leverage ratio is expected to move under 1x and toward our maintenance net leverage target of less than 0.0x. We believe low leverage and ample liquidity gives us significant flexibility to drive long-term shareholder value through disciplined capital allocation.

This brings me to our go-forward capital allocation priorities. First, organic investment, customer-aligned R&D focused on our highest return opportunities with the clearest visibility. Second, portfolio optimization, applying a disciplined framework for opportunistic, capability-driven M&A, prioritizing strategic fit and the strongest returns and staying mindful of our history. Third, balance sheet strength, preserving the flexibility to act on operational and strategic opportunities as they arise.

Let me ground this in a recent example, our acquisition of Semtech Photonics Products or HieFo in March of this year. Semtech Photonics Products or SPP, expands our AI connectivity portfolio into photonics and adds domestic indium phosphide manufacturing expertise that would have taken us far longer and would have been far more costly to develop organically. The economics are attractive. SPP's photonics portfolio is expected to generate data center level gross margins with clear R&D and go-to-market synergies. And the transaction was appropriately sized to preserve our balance sheet strength with capacity expansion, both what we've spent and what we expect to spend within our liquidity framework. This is the model going forward, opportunistic, capability-driven acquisitions that strengthen our core franchises, not scale for its own sake.

So allow me to translate all of today's presentations into our long-term financial framework. Our accumulated efforts in revamping customer engagement, the concerted R&D investments in high-return, customer-aligned projects, our continuous portfolio optimization, the multiplier effect from stronger employee engagement and assurance from our financial transformation together form the foundation of the framework we are calling Semtech Excelling.

We expect revenue to grow at a 38% CAGR from FY '26 to FY '30, reaching $3.2 billion. I want to be precise in the base and that we exclude $166 million of FY '26 cellular module revenue, the business that we expect to divest in Q4.

Data Center revenue is to compound at 75% to $2.1 billion for FY '30. As Imran mentioned, we also expect data center revenue to more than double in FY '27 versus FY '26 and then more than double again in FY '28. Imran showed you why. FiberEdge's differentiated performance, CopperEdge's differentiated performance, Photonics with differentiated performance just starting to contribute to revenue. Content opportunity per module growing at over 10x, a strategic supplier with whom we've had a long-term supply relationship aligned to supporting multiple definitional customers and all across the SAM that we expect to grow from $3 billion to $15 billion.

LoRa's revenue is expected to compound at 30% through FY '30, supported by capability enhancements and expanding market verticals, including smart building, security, drone control, edge AI connectivity and our emerging partnership with Amazon Sidewalk. We expect steady growth from the rest of the portfolio above market growth rates and contributing cash flow and contributing operating leverage. Alongside revenue growth, we expect adjusted gross margin of 70% in FY '30. For this quarter, excluding the cellular modules business, we expect gross margin of 64%. That remaining 6 points are contributed through mix with Data Center and LoRa growth and growing scale of our business. Adjusted operating margin for FY '30 is expected to range from 48% to 50% with R&D investment as the primary variable. And of course, incremental R&D will follow our same framework of customer-aligned projects. We also expect our growth to result in healthy cash generation with free cash flow as a percentage of revenue to be 35% in FY '30. Together, that drives adjusted diluted EPS CAGR of 64% to 66% or adjusted EPS at $12.40 to $13.

We've assumed a 20% non-GAAP tax rate based on geographic mix of income and 104 million weighted average diluted shares outstanding. To close, this is our model to compound shareholder value, winning in high-growth markets by aligning investment in our customers' needs, increasing portfolio value by building a more differentiated, higher-margin portfolio, expanding earnings power by turning growth into stronger earnings, generating durable cash flow by converting those earnings into durable and sustainable free cash flow and then reinvesting with discipline to deploy capital to maximize returns. Quality top line growth, margin expansion, cash generation, disciplined reinvestment. That's how we intend to compound shareholder value in the years ahead. Thank you.

Please welcome back to the stage, Mitch Haws.

Thank you, Mark. At this point, we will open the floor to questions. I would like to invite members of the senior leadership team to join me. Just a couple of rules of the road. There are mics in the audience. Before you pose a question, please use the mic for the benefit of those on the webcast. And if you could, please state your name and the firm you're associated with. With that -- we do have 2 additional members of the team that haven't presented today. So I thought it might be good if they just made a quick introduction. Jason?

Sure. Jason Green, I'm the Chief Commercial Officer. So I lead the customer-facing teams and the go-to-market effort.

Asaf?

Hi, everyone. My name is Asaf Silberstein, and I'm the Chief Operating Officer in the company.

So the first question, Chris, far away.

Before we start the Q&A session, I have a very important announcement. As we were preparing for the Analyst Day and Mitch was preparing to be a grandfather. Great day today. Mitch became grandfather -- congratulations.

Thank you.

A beautiful grandson was born in at 2:30 a.m. this morning.

Can we call you papa?

Some of you know that Mitch Jr., Thomas, is also IR executive. We have nothing to -- more than that we expect that Thomas Jr. is going to be an IR executive as well. Sorry about that, Chris.

Far more important than anything I have to say. And I know Thomas and I guess congratulations to Thomas as well.

Thank you, Chris.

That's incredible. And with your youthful appearance, I would not have expected a grandfather. Thank you for a great day and really talking about all these opportunities that you have ahead for you. I have received perhaps a couple of questions about the long-term model and the way it's structured. And the way that growth starts out really robust. And then I think in the last 2 years, there's an implied decel in the growth to much more of a market rate. And so I think there's -- I think people would like to know basically what's in that assumption in the back half? Is it just uncertainty? Is it something you're seeing in your own businesses and the portfolio and the way it's set up? Is it just conservatism? And what would it take to beat that?

Yes. So thanks for the question, Chris. So our long-term growth rates in those outer years are grounded in market CAGR assumptions, market growth assumptions, market levels. And if the market can grow faster and the market is bigger, we'll be there to capture those opportunities without a doubt. But let me just ground our model and the framework we use to develop our model. Semtech has an established long-term financial planning plan and the way we develop our numbers. We applied that framework. We applied those same processes and procedures in developing our -- the long-term model that we shared today. We didn't use anything different. We just maybe updated some of the inputs based on the latest market data. But that also grounds our belief, just maybe applying to Data Center, right?

In FY '27, we expect Data Center revenue to more than double. That's grounded in backlog. FY '28, our growth rates more than doubling again is grounded in backlog and then maybe layering in design wins that have yet to ramp. We also layer in a bit of customer engagement there, right? Where are our customers headed will align to where we're headed. And then, of course, supply assurance, right, what we can supply. And I think the work that Asaf has done with Tower Semiconductor, I think solves for that particular variable pretty well. In outer years, we ground our expectations and what we see in a market CAGR, the market CAGR that we presented today. So that leads to some variability in outcomes. But absolutely, if that market CAGR grows beyond what we presented, you can expect that our growth rates will expand beyond what we presented today as well.

Thanks, Mark. Next question. How about Quinn Bolton.

Quinn Bolton with Needham. I wanted to follow up on sort of the long-term outlook. You've given us a Data Center target of $2.1 billion. The current business is mostly scale-out or optical module based. You've talked about NPO. As you look out to 2030, how much of the $2.1 billion do you think could be generated by NPO, CPO sort of new applications coming to market versus growth in the traditional optical module business? And then I've got a follow-up for Mark.

Yes. So Quinn, thank you for your question. We will not be able to provide the detail from the different parts of the data center portfolio contribution. But it's safe to say that all of them is contributing very nicely to the future growth. Again, on the outlook, today, we provide FY '30 simply follow the framework as Mark outlined. It does not reflect any, say, we sense some deceleration for future years. It's simply because we have the backlog to support this year and next year's conviction. And then as we go in the future, we will provide more update. And if the market continues to go at this way, we absolutely will leave no opportunities unexplored. So we're going to capture all the growth opportunities.

Quinn, you had a follow-up?

Just a question. You went through your capital priorities, obviously, continuing to invest in the business. But if you hit your revenue target and your free cash flow target of 35%, but be generating over $1.1 billion in fiscal '30. What do you think the opportunity is for share buybacks, dividends? Just any sort of capital return thoughts as you get out to that model?

So I'll frame our capital allocation priorities. First of all, investing in R&D in these high-growth projects that we've been able to identify, and we believe we'll continue to identify through really strong customer engagement. That will be the first use. And I think we've demonstrated a great return, both top line growth and bottom line return from that investment in R&D. M&A is another area. Portfolio optimization is a continuous process. And just looking back at our framework, what have we been looking at in M&A? On the acquisition side, capability-driven acquisitions that target kind of adjacent markets, kind of fill out the portfolio. We can expect to see more of that perhaps at various scales. And then third, balance sheet strength. That means having a fundamentally strong balance sheet, getting to that net leverage ratio target over time, the maintenance net leverage ratio target of 0.0x.

Okay. Joe?

Joe Moore from Morgan Stanley. Thank you for the great presentation, and I'm excited to hear your grandson is continuing in the tradition of semiconductor IR down the road. Can you talk about active copper? You talked a lot about redriver opportunity persisting for several years. What's the adoption there? Are you seeing -- is this slice of it, maybe a little smaller than I might have thought a couple of years out? Is that adoption curve looking like what you thought? And is it just that optical is a lot stronger? Is that why it's a small piece?

Imran, do you want to address that?

Sure. I can do so. Thank you for the question. We're less focused on the format, if you like, of whether it's active copper or whether it's a linear redriver on a line card. And I would encourage you to think of it the same way. There's no sense in getting hung up on how -- what is the format of these links. What we need to understand and what I hope came through in my presentation is that these links need a boost. They need a tailwind, and that's what linear redrivers provide. And in all of these applications going forward as line rates increase, we expect to see exactly that. Some of them will take the form of ACC. Some of them will take the form of a flyover cable with a redriver at the end. Some will be a redriver instead of a retimer on a line card. Thank you for the question.

How about Sean in the middle here?

It's Sean O'Loughlin from TD Cowen. Thanks for the day of presentations and congrats to Grandpa over there. Maybe one for -- I don't know whether this is a Hong question or an Imran question. But when we think about the Semtech Photonics portfolio and -- or the Semtech Photonics products portfolio, that's PPP. Right now, I think that you've commented that the gain chips are so wildly profitable that you're continuing to make those and the shift to lasers will come later. And when I think about where there are bottlenecks in the industry, a lot of those are in whether they're pump lasers, whether they're EMLs, whether it's CW lasers. I guess you were clear on what the road map looked like. But I guess what is different about your portfolio that makes maybe those products that you highlighted easier to tackle first than maybe other parts of the broader laser market, if that makes sense?

Can I answer that. Great question. Now it's -- I'm awfully familiar with the laser and the capability and the differentiation we developed at the time. It's the same high-power lasers at the time was used to serve the HFC hybrid fiber coaxial market. For those of you who have been around for a while, that was a big market. Today, if you use Xfinity, I would say, over 80% of the chance that network is powered by the lasers we designed and offered at EMCORE days. The differentiation for lasers is high conversion -- power conversion efficiency, about 10% higher from 30% to 40%. That is significant.

And also the beam profile was designed in such a way that easy to couple into waveguide or fiber, you get a less loss and more laser to get in and a very low noise and very low bandwidth. What does that translate into? We're not going to be competing with the commodity type, like a $3 per 30-milliwatt lasers for optical transceivers with more high value-add parts, for example, gain chips providing the key elements in tunable lasers for scale across applications. That's a foothold. That's our base. We continue to strengthen that base by expanding customer base.

The second, driving the adjacent, which is Coherent-Lite that's going to be emerging scale across applications. Our laser is ready. Our lasers are in qualification cycle. And then we're selectively engaging on optical transceiver applications. And we're not going to be using 8 lasers to serve 1 optical transceiver. We plan to use one, engage with our customers using our very high-power laser 150 to 200 milliwatt to 250 milliwatt depends on their loss to have 1 laser serve 8 channels. That's more value add and then developing the OCI CPO scale-up lasers. So that has been our strategy. This coincides with our capacity expansion plan very nicely. So different opportunities layering in at a different time. We understand the horizon of the different applications, and we plan accordingly.

Harsh?

Harsh Kumar, BMO Capital Markets. Mitch, congratulations on becoming a grandfather. For my question, Hong, I wanted to ask you about the importance of your deal with Tower semi. What does it bring to you that HieFo doesn't have? Maybe you could contrast and help us understand what you're getting out of that deal and what kind of capacity implications it has? And then I had a follow-up.

Thank you, Harsh, for the question. So if we open the hood like what we did today on an optical transceiver or NPO or XPO construct, the key components in play are PMDs with the TIA and drivers and then photonic component, you got lasers and modulators and photodiode. We don't do DSP. We're not planning to do DSP as the market is moving into this trend of more DSP less for scale-up and scale in and even in many part of the scale-out network. So we look at that laser, we have a strong foothold, and that can be the base to expand. Photodiode, we just added design capability, as I alluded, we plan to demonstrate our 300-gig photodiode in the upcoming OFC. And okay, you can say, "well, what about the modulators?" We're working with the partners. And so we have many ways to play from that angle as well. Our goal is to work and co-optimize those components, provide customers with a reference design and known good chipsets. So this will help them to accelerate the new product to market.

My follow-up. Hong, I wanted to ask you about NPO versus CPO timing of CPO in your opinion. And then there's always this push and pull between investors when we talk to them, they talk about, hey, there's a risk that the switch ASIC guys will try to take more of the share. What would you say to something like that? And also maybe you could in the same vein and address the photonics fabric that might be coming on a couple of years out?

Yes. So this area, I happen to know something we've been working on the CPO since 2018 at Intel. We were a little ahead of time. We delivered working CPO in early 2020 and was fully prepared to showcase at OFC and then COVID and then canceled. But the program that slow rolled primarily is not because of the technology. It's because of application. There are no killer app which need that much bandwidth, more importantly, bandwidth density. AI changed the game. So now there's so much data cranking out from XPUs and people talk about bandwidth limitation, connectivity wall and also the power wall because you use DSP-based solutions, you're getting the signal, bring them all the way to the front panel, you burn so much power. The connectivity consume too much power in their eyes, and that has been impediment for them to monetize more compute revenue through the power. So we are putting in the front line in solving this issue, these challenges.

CPO is going to happen, and there are no fundamental scientific barriers, but more process development maturity. But in between, there is an open system CPO, you can view it that way, that's called NPO. Near package, you don't get 100% benefit, but you get 85% there in power savings. That -- but you leave the socket open and give people the flexibility you're dropping in different type of XPUs. So I think the NPO is going to happen. The volume will start pretty much end of next year. And we are at the right place. As I said, we have all the key IPs, which really basically NPO is a highly packed LPO. The TIA, the drivers we develop for LPO and proven and that can be used in NPO with the right form factor to match the packaging need. As for CPO, is going to happen. That will be the holy grail.

In the past, the concern was mostly about serviceability. So when we developed that at Intel, Arista came by several times saying, Geez, if this thing doesn't work, who do I send to? Send to you, send to Broadcom for switch or send to Celestica for manufacturing. They don't even know how to deal with the fibers. Now you have 1 neck to grab. NVIDIA, for example, they provide the racks, you can hold them accountable for the rack performance. Google provide the rack, you can hold them accountable. So that's no longer an issue. And CPO, again, is going to be happening, but happening after NPO.

And maybe to the second part of your question, I'll just add regarding the competition and as everybody goes in the space, I can share with you that customers really appreciate our focus in on the analog connectivity space. We really come across them as trusted partners rather than potential competitors down the road as it gets to CPO and beyond. So I think that's an important point to consider.

Thanks, Jason. Rick in the back there.

I'll add my congratulations to Mitch. Rick Schafer from Oppenheimer. Hong, I've got a question on LoRa. I know we've been talking a lot about data center, but I'm trying to get a sense of how big that market could be. I mean you've taken that from a no-growth business in the last couple of years. It's running $0.25 billion, I think, now, growing 60%. I know you talked about long-term CAGR around 30%. But I look at just something as simple as Amazon Sidewalk, which seems like it has the potential to be as big or bigger than what your run rate is now with LoRa. And I don't know if you agree with that, but I'm trying to get a real sense of basically how big that LoRa opportunity is or that TAM is.

Yes. Rick, thank you for that question. You're right. And you and I, we talked -- so LoRa revenue was almost at $150 million plus/minus on an annualized basis. This year, we're going to break that shatter that glass ceiling to $250 million, $260 million a year. The growth is primarily driven by unleashing new application verticals. You mentioned about Amazon Sidewalk. We actually put in a very moderate number in that in our multiyear plan. That can be a huge. It's a consumer application, and we rely on their Amazon scale to scale up. But what we have the full visibility and more control over is the value-added application across all different application verticals.

One application we did not get a chance to talk a whole lot is the edge AI applications. LoRa is known to provide a robust, wide range, low-power and super sensitive connectivity solutions. But now we are adding bandwidth in it. In Gen 4, we have 2.4, 2.6 megabit per second compared to 100 kilobit per second. In Gen 5, Mark talked about start by our Switzerland design center. I was there, too. So we're so happy to hear that our innovators are able to bring that bandwidth to 6.4 megabit per second. Gen 5 with even more sensitivity by 2, 3 dBs. 6.4 megabit by any imagination can be called broadband, but it's getting there. And what that enables is the high-fidelity video and audio applications in real time. Imagine you have the robust coverage, long reach, low power, adding high bandwidth that can unlock a whole range of new applications. So you can tell, I'm really excited about the future of LoRa. There's just a new range of many applications that give us the confidence and conviction that we will be able to grow with a CAGR of 30%.

Right here, Craig?

Yes. Great presentations, guys, very informative. Mitch, congratulations. Hong, I wanted to start with a clarification. I think it was your Slide 21 that had that very attractive-looking bar chart with content going above $100 out in the future when we integrate laser. And the question is a clarification. To what extent in the $3.2 billion revenue model for fiscal '30, $2.1 billion for Data Center, do we realize that kind of content gain? Or is that really beyond the forecast period in fiscal '31 and beyond?

Simply put, a very moderate amount, and we did not really put in even close to what we believe the potential can be in our model. It does need to do a lot to make it happen, but our path is clear. Our opportunity is there. And we are -- all we need to do is focus on capture those opportunities.

Fantastic. Very helpful. Mark, the follow-up is to you. I didn't want to neglect you up there on the panel. The 70% gross margin bogey for fiscal '30, you said that there was scale and mix that were drivers there. Can you talk a little bit about each of those and the relative contribution and then the extent to which that's more linear or maybe back-end loaded in terms of getting there from 64%.

Yes. Thanks, Craig. I call it less of a bogey and more of a target that's grounded in the modeling that we do for our long-term financial plan. Again, we don't apply a different framework for how we generate these numbers. We've done 5-year plans for quite some time now. That's just part of our standard process. Now in terms of getting to that 70%, it's 6 points, right? We're at 64% ex-cellular modules for this quarter. And you can see where the business is growing within data center and LoRa. I think most of the people in the audience are familiar with gross margins in this area, but let me just reinforce, these are among our highest gross margin products, let's call it our highest gross margin products in our portfolio. So with that growth rate, absolutely, that's driving towards a higher gross margin.

Scale, of course, we can't ignore it either, right? You've seen scale growth. You can see how the business is growing just in the past few quarters, apples for apples. Scale has definitely contributed. But I'd say that the driver that move towards this mix of data center and LoRa is a really powerful tailwind. But we expect that tailwind to continue to exist, to continue to grow. And look, we based these long-term figures on a SAM assumption. The SAM assumptions will go up. Absolutely, we'll be there to capture them. That will have intended impacts, positive impacts in our gross margin as well.

Take the mic, Tristan.

Tristan Gerra. A quick question on FiberEdge. In your slide, you talked about meaningful market share gains and you're now slightly above 50%. Maybe that's a question for Imran. What's your expectation of share gain going forward? Is that embedded in your guidance? Or do you think you stabilize at current levels? And also, if you could talk about the competitive landscape for ACC at 400 gig per lane.

Imran, you want to answer that? I can chime in later.

Sure. And Jason, you might want to jump in here as well. Thank you, Tristan, for the question. As I mentioned, we grew our share from 18% to 50%. And we have no intention of stopping there, not in any of the products. Our ambition, our aspiration is to deliver what the market wants with the best possible performance. And we believe that we're going to be very successful in doing so. Jason, maybe you want to.

I would just add that we have more visibility than we've ever had before, and that's because we're closer to customers than we've ever been. And so when we have the customer engagement that we're continuing to grow, having the success with 100 leaves us in a great position. We feel great about where we are with 200 gig per lane.

And as to your question, Tristan, on ACCs, again, as I mentioned earlier, we stopped looking at those opportunities separately like that. For us, the channel is the channel, whether it ends up being a cable with -- in an ACC format or whether it ends up being a trace on a PCB, it doesn't really impact our business, if you like. What we're focused on is delivering those linear redrivers at ever-increasing rates with the best performance out there. And as I mentioned in the talk, right now at 336, we seem to be far ahead. Okay.

One more question on this side.

This is Ali from Irrational Analysis. I want to ask about the black magic of analog design. If we take, for example, your linear equalizer, which is really a single-stage CTLE that has only peaking gain adjustment, it's probably 5 to 10 transistors for the amplifier and 4 to 6 variable resistors or capacitors with 3 bits resolution each for tuning across process variation. And at a schematic level, this is a very simple design. However, it's a very high gross margin part. It's a very valuable part. So there's quite a bit of difficulty in that design. So I'm curious where does the difficulty come from? Is Semtech perhaps significantly better than the competition at dealing with silicon germanium process variation? I just want to know some more color on that.

Imran, you want me to take this one?

And I have a quick follow-up.

Ali, that's a great question. And I'm really encouraged actually that there's somebody here talking at the transistor level. That's fantastic. I think we can take that question offline. I have my engineering colleagues at the back there, who I'm sure would love to discuss that with you in more detail.

Okay. Can I ask my quick follow-up question, which will also probably go offline. Normally, in a high-speed SerDes on the CMOS, there is a 2- or 3-stage CTLE that has peaking adjustment, peaking location adjustment, a mid-band adjustment and resistive degeneration that can attenuate from, let's say, 0 to 2 gigahertz. And my understanding is Semtech's discrete CTLE along with the competition do not have these adjustments only peaking. And in my experience, it's very, very helpful, particularly resistive degeneration to tune a link and get it working and get that last 1 or 2 orders of BER improvement. And I'm curious why Semtech has not implemented that. And perhaps if one of the many R&D spend topics you've discussed, you maybe have that in the pipe. So I'll see you afterwards.

Wonderful. Thank you. Great. We look forward to it.

That's probably a good point to end the Q&A session. Running out of time. Maybe I'll turn the clock back to Hong for some final comments.

So I wanted to thank everyone for coming to this great event. I know it's a long week. We have been walking on the floor at OCP. I know the market has not been very kind. It can be an exhausting week. Thank you so much for sitting through the presentation. If the key takeaways I want you guys to go walk away with, this is a generational opportunity. And connectivity is a bottleneck. And connectivity need high bandwidth, low power, low latency solutions that favors to analog as the market is moving toward us. We just drive the inflection and the company has a strong culture foundation has a comprehensive product portfolio, had the application -- have the execution track record. We are very excited and to really take on the challenge to outgrow the market and deliver the best results and deliver the best return to our shareholders. We have proven that we have embarked the journey of transforming, and we have been showing you the early proof points, but the best is yet to come.

Thank you very much. You still have time to roam around and engage with our senior leaders and some of our Board members are here as well. So feel free to interrogate them as well. And thank you, and have a great day.

Good day, and thank you for standing by. Welcome to Semtech Corporation's Second Quarter 2027 Earnings Conference Call. [Operator Instructions]. Please be advised that today's conference call is being recorded. I would now like to hand the conference over to Mitch Haws, Senior Vice President of Investor Relations for Semtech. Please go ahead.

Thank you, and welcome to Semtech's Second Quarter 2027 Financial Results Conference Call. Participants on today's conference call are Hong Hou, President and Chief Executive Officer; and Mark Lin, Executive Vice President and Chief Financial Officer.

Before we begin the prepared remarks, I would like to highlight upcoming investor events, including the Citibank Global TMT Conference on September 8 and the benchmark TMT and JPMorgan Rising Tech Leaders Forum, both on September 10 in New York City. In addition, we hope you'll attend our investor event in San Jose on October 15 and during which we'll provide an in-depth overview of Semtech's strategy, differentiated technology portfolio, key growth opportunities and long-term financial targets. The event will also feature panel discussion moderated by Morgan Stanley with industry luminaries from the 650 Group, Meta and General Catalyst.

A question-and-answer session, product demonstration and opportunities for in-person attendees to engage with members of the Semtech management will also be part of the agenda. Today, after the market closed, we released our unaudited results for the second quarter ended July 26, 2026, which are posted along with an earnings call presentation to our Investor Relations website at investors.semtech.com.

Today's call will include various remarks about future expectations, plans and prospects, which comprise forward-looking statements. Please refer to today's press release and see Slide 2 of the earnings presentation, as well as the Risk Factors section of our most recent annual report on Form 10-K for a number of risk factors that could cause our actual results and events to differ materially from those anticipated or projected on today's call. You should consider these risk factors in conjunction with our other forward-looking statements. We will refer primarily to non-GAAP financial measures during today's call, and we'll also be referring to results for our second quarter of fiscal year 2027, unless otherwise noted.

Please see today's press release and Slides 3 and 4 of the earnings presentation for information regarding notes on our non-GAAP financial presentation. The press release and earnings presentation also include reconciliations of our GAAP and non-GAAP financial measures. With that, I will turn the call over to Hong.

Thank you, Mitch. Good afternoon to all of you joining today. The Semtech team executed exceptionally well this quarter, delivering righter revenue across our key focus areas earnings leverage that continue to outpace revenue growth and significant progress on portfolio optimization. Revenue was $342 million growing 33% year-over-year, and we delivered strong operating leverage with earnings per share of $0.71 in 73% year-over-year, more than twice as fast as revenue growth. We are at the center of one of the most significant infrastructure build-out in history, and our portfolio plays an essential role.

We are well aligned with the ramp to 160 in complementing 800-gig growth and demand signals that we are strengthening across every part of data center portfolio, copper, fab and photonics. We expect this momentum to carry through the second half of the year and into fiscal 2028. We are also reshaping Semtech with purpose. The announced sale of our several module business is a significant step in our portfolio optimization, allowing us to more sharply focus on our core growth areas. We are drilling in our focus areas, sharpening the portfolio and driving operating leverage with the same goal in mind, building a predictable, high-margin and high return business.

Now let me move on a discussion to our end market. Infrastructure net sales were $124 million, up 25% sequentially and 69% year-over-year driven by outstanding performance in our data center business. Data center revenue was a record $100 million, up 39% sequentially and 91% year-over-year. Supported by continued strength in 800 gig, 1.6 copper edge and the start of our 1.60 fiber edge rep. Our FiberEdge TIA and driver solutions remain an exceptionally strong demand. And we continue to deepen our engagement across all the leading hyperscalers. We are now designing to every module provider in our target markets, several on a sole-source basis, a reflection of technology differentiation and the supply availability we bring across both fully retimed and linear architectures.

We are also seeing increasing engagement from a broader array of customers on emerging technologies like NPO and XPO. As the networking ecosystem looks to us to align and help define the next generation of high-density, low-power optical architectures in our shared technology road map. On copper edge, we believe our leading [ EcoLaser ] solutions are the de facto industry standard. Copper edge product up to 1.60 are solutions that are ready for volume deployment. We are currently engaging across a number of hyperscalers in cable and onboard applications and the design win phase at our bandwidth up to 3.2.

Thanks to leader equalizers compelling advantage in linked margin performance and power savings. Based on strong market demand and a design win momentum, we expect continued revenue growth of 1.6 portfolio with the fiber edge expected to exceed market share by the end of the fiscal year and a coverage already taking the lion's share of the leader Equalizer market. We have made excellent progress in our photonics portfolio. [indiscernible] our customer base in both game chips and high-power CW lasers addressing both high-speed transceivers and CPO scale-up applications. Feedback from customer evaluations of our high-power CW laser for procure light and 1.60 transceiver applications has been very positive, citing differentiating over temperature performance and power efficiency.

We expect revenue contribution of CW lasers for transceivers to start in the first half of fiscal 2028. We're also pleased to help route onboard auto diode design resources added by an industry leader expanding our photonic portfolio to PD raise in the near future. Our combined PD and PIA design team has already engaged with the key customers and we expect to deliver to optimized high-performance solutions. Our photonic portfolio now spans game ship type lasers semiconductor optical amplifiers and high-speed factoids for scale up, scale out and scale of breast data center connectivity applications.

With this expanded portfolio, we are positioned to develop new prescribers and drive content per transceiver from high single-digit dollars to high double-digit dollars as the industry transitions from 800-gig to 3.20, cementing our position as a true solution provider. Our capacity expensing plan, our team executed very well, securing equipment deliveries for this fiscal year and acquiring clean room space to fulfill strong customer demand.

In less than 6 months, we completed a series of photonic acquisitions, procured fab equipment expanded clean room space and onboarded exceptional management and technical talent. We have established a solid foothold in the photonics space and set path for strong future growth. Given record backlog we carry into the third quarter, we project a 45% sequential revenue pie in data center representing approximately 160% over the same period last year.

We expect accelerating year-over-year periods into fourth quarter and continued momentum throughout fiscal 2028. Now moving to our high-end consumer end market. Net sales for Q2 were $39 million, up 2% sequentially and down 5% year-over-year. Our TV business grew sequentially and remains very resilient in light of memory constrained price across the industry. Revenue growth continued to benefit from our strong share at premium brand handset manufacturers where we are expanding our content per device. Third switch our newest [ certain ] protection solution is opening a new layer of TVS opportunity, addressing a GAC as a rugged mobile devices and high-performance portable systems cost towards more demanding power and reliability standards.

Our per se capacity sensor design and pipeline continues to drive its specific absorption rate smart variable and other consumer applications, expanding with the lead customers on a broadening range of applications. The combined capacities and force sensing offerings elevate our value proposition, strengthen customer retention and are pulling through sensors and TVS sales within the same customer base. We expect our design win pipeline to support the long-term growth for this business. Now moving to our industrial end market. Q2 industrial net sales were $179 million, up 16% sequentially and up 25% year-over-year, driven by another record quarter for LoRa.

LoRa enabled net sales were $58 million, up 31% sequentially and up 58% year-over-year, another all-time record. A lower Gen 4 platform will LoraPLUS other RF protocols continues to gain market traction, and we expect it will be a key driver for the future growth. Gen4 also delivers a due brand capability and expand data throughput to 2.6 megabit per second, while preserving the sensitivity multi-protocol flexibility and ultra-low power consumption that defines a lower advantage. This feature set enables new class of edge AI applications while maintaining the long battery life and the extended reach that our customers depend on and opens up incremental application verticals within smartphone and security. We also continue to see LoRaWAN an expansion into new use cases.

In public safety, sensors can now transmit high fidelity audio for AI-based verification rather than simple alert. And in the industrial environment, our work with industry leaders demonstrates how LoRaWAN and edge AI together enable predictive maintenance at a level of the detail that legacy low-power sensors could not support. Amazon side will continue to build momentum.

Following Rain's launch of a new line of LoRa-based sensors in the U.S., [indiscernible] is now expanding internationally, starting with Canada and Mexico, with Europe, Australia and Japan expected to follow. This is a meaningful step towards mass-market consumer adoption at Amazon's scale. Together, our 3 pillars LoRaWAN for industrial and commercial deployments, LoRaPLUs with multi-protocol flexibility for smart home and security and the Amazon Sidewalk for mass-market consumer applications continue to create a solid framework for Pros.

We project another all-time high for LoRa revenue in Q3 with a growth of about 15% sequentially, equating to year-over-year growth about 65%. Our IoT systems and connectivity business recorded Q2 net sales of $98 million, up 11% sequentially and year-over-year. Our air link routers saw strong new business activity across mission-critical applications, driven by drilling engagement with the national carrier partners on 5G stand-alone network slicing. This momentum was reinforced by our RX 400 and EX 400 5G RedCap routers moving into full-scale production this quarter. with wins continue to convert into shipments across a broad range of customers. We also continue to invest in AirLink software platform to provide new security and device management capabilities.

These capabilities are giving missing critical customers greater visibility and control as they manage larger, more complex deployments reflecting our broader commitment to software R&D as a way to deliver more capability and value to our customers over time. In summary, our second quarter results reflect significant progress in Semtech's transformation, including a strong winning culture. But to be clear, the progress we are making is just the foundation, not a finish line. Our priorities for fiscal 2027 remain the same and are straightforward.

First, supporting our unprecedented backlog and the drill opportunities, we are actively securing incremental capacity for fiscal 2028 and beyond. Second, intensifying R&D investment to support customer technology road maps in a rapidly advancing market and adding new drill drivers, specifically in solution offerings for lasers, photo dials drivers and TIAs for 3.2 coherent light, XPO, MPO and the CPO applications. And the third, continuing portfolio optimization. We see this as a continuous journey and there is more work ahead of us as we reshape Semtech. This is such an exciting time for Semtech. The business is just starting to inflect and the opportunities ahead has never been more compelling.

With that, I will turn the call over to Mark for additional details on our financial results and our third quarter outlook. Mark?

Thank you, Hong. For Q2, we recorded our tenth consecutive quarter of net sales growth with record net sales of $342 million, above the high end of our outlook range. Net sales grew 17% sequentially and 33% year-over-year, reflective of leverage in our operating model. We reported adjusted diluted earnings per share of $0.71, which increased at over 2x the rate of net sales growth on both a sequential and year-over-year basis. Net sales trends by end market, reportable segment and geographic region are included in the accompanying earnings presentation.

Adjusted gross margin was 54.5%, up 150 basis points sequentially and at the high end of our outlook. Total semiconductor products gross margin was 62.8%, up 210 basis points sequentially and above the high end of our outlook, reflecting particularly strong contribution from 1.60 FiberEdge and copper Edge and continued growth from our LoRa portfolio. We announced the signing of a definitive agreement to divest our cellular module business, which is recorded as held for sale on the Q2 balance sheet.

To facilitate comparability for our go-forward business, we added an adjusted gross margin disclosure in our earnings release and earnings presentation that excludes the held-for-sale business. Excluding the [indiscernible] business, Q2 adjusted gross margin was 59.7% or 520 basis points above consolidated gross margin, reflecting the magnitude of the structural shift on top of the 150 basis points of sequential consolidated gross margin improvement. We expect to provide a gross margin outlook, including and excluding the cellular module business until the close of the divestiture which is expected to occur in the fourth quarter of the current fiscal year.

We also expect the transaction to be EPS neutral on a non-GAAP basis. Adjusted net operating expenses were $103 million, below the low end of our guidance range, reflecting timing of project-related expenses. Demonstrating the operating leverage in our business, a number of metrics were favorable to the high end of our guidance range, including adjusted operating income of $84 million, adjusted operating margin of 24.4%, adjusted EBITDA of $91 million and adjusted EBITDA margin of 26.6%. Collective of capital structure changes, Semtech remained in a net interest income position in Q2. We recorded adjusted diluted earnings per share of $0.71, above the high end of our guidance range. up 39% sequentially and up 73% year-over-year.

Operating cash flow for Q2 was $69 million, up 90% sequentially from $36 million and up 55% from $44 million a year ago. Free cash flow for Q2 was $61 million, up 119% sequentially from $28 million and up 48% from $42 million a year ago. CapEx was 2% of net sales and includes expenditures to grow fab capacity supporting gain ships and CW lasers. We expect CapEx to grow as a percentage of sales, but to remain manageable and generally be below 5% of net sales. Though timing of construction and equipment delivery could increase as percentage slightly on a single quarter basis.

Our Q2 ending cash and cash equivalents balance was $204 million and the principal amount of debt was $503 million and net leverage ratio was 1.1. Now turning to our outlook for the third quarter of fiscal year 2027. We currently expect net sales of $410 million, plus or minus $5 million, up 20% sequentially and up 54% year-over-year at the midpoint, with growth expected across each of our segments. We expect net sales from our infrastructure end market to increase sequentially with projected sequential data center growth of 45% or 160% year-over-year with continued strong contribution from our 800 gig portfolio and a meaningful ramp in 1.60 copper edge and FiberEdge.

We expect net sales from our high-end consumer end market to increase, benefiting from seasonal trends, market share gain in our TVS products and contributions from our sensing portfolio, we expect net sales from our industrial end market to broadly grow with or revenue increasing about 15% sequentially and 65% year-over-year. Based on our expected product mix and net sales levels, we expect adjusted gross margin to be 58.3%, plus or minus 100 basis points. At the midpoint, this equates to an increase of 380 basis points sequentially and 530 basis points year-over-year.

Our gross margin outlook, excluding the cellular module business is expected to be 63.9% at the midpoint, an incremental 560 basis points from the midpoint of the consolidated adjusted gross margin outlook. Adjusted net operating expenses are expected to be $112 million, plus or minus $3 million. Included in this outlook is increased R&D spend to accelerate time to market on key data center projects along with SG&A that declines as a percentage of revenue. We have demonstrated strong returns on our R&D investment and believe we remain prudent on SG&A spend.

This results in consolidated adjusted operating margin at the midpoint of 31%, up 660 basis points sequentially and up 1,040 basis points year-over-year. Adjusted EBITDA is expected to be $134 million plus or minus $4 million, resulting in adjusted EBITDA margin at the midpoint of 32.8%, up 620 basis points sequentially and up 930 basis points year-over-year. We expect adjusted interest and other expenses net to be approximately $0.5 million. We expect an adjusted normalized income tax rate of 18%, reflecting geographic mix of income. These amounts are expected to result in adjusted diluted earnings per share of $1.05, plus or minus $0.03, up 48% sequentially and up 119% year-over-year at the midpoint, more than 2x revenue growth based on an expected weighted average share count of 99 million shares.

I look forward to providing our financial framework and multiyear outlook at our upcoming investor event on October 15. We expect the framework will highlight the operating leverage in our business model, namely increasing gross margin, reflecting strong contributions from data center and LoRa. Operating margin that grows with scale and with disciplined spend in G&A hoping to support R&D investment and a structural shift in margins following the cellular model divestiture all of which are expected to support strong EPS, EBITDA and cash flow metrics. With that, I'll turn it back to Mitch.

Thank you, Mark. We can now turn the call back over to the operator for the question-and-answer session.

Our first question is from Quinn Bolton with Needham & Company.

Congratulations on the strong results. Hong, you mentioned needing to go out and secure capacity for fiscal -- sorry, calendar 2028 and beyond. But the data center business, I think guiding up 160% year-on-year in the third fiscal quarter. It sounds like it accelerates in the fourth fiscal quarter. How are you feeling near term about capacity in your ability to support continued upside in the data center business? And then I've got a follow-on data center question.

We anticipated a very rapid data center revenue growth. We started about 1.5 years ago, and thanks to that work, we were able to have enough capacity in the near term to support the customer ramp and also some drop in orders. So that allow us to expand our market share. Now with the strong booking momentum and record backlog, we see the capacity we have secured may not be enough in supporting the FY '28, especially second half of '28. So working with our manufacturing partners, both for front end and back end back end means that OSAT from testing to packaging and disease testing, working with the manufacturing partner to increase the capacity. The great news is that we have the financial capability to work with our partners to jointly increase the capacity allocation to some tech

I guess maybe quickly, just Hong, does that -- would you anticipate that requiring wafer pre-purchases or any kind of similar prepurchases of back-end capacity. And then my follow-on question was just -- it seems like there's growing discussion of NPO solutions across the ASIC landscape. And I think even at the largest GPU provider as we look into the next 12 to 24 months, can you just give us a brief outline of how Semtech is positioned to support the NPO market as it develops?

Yes. Thank you. So yes, the increase of capacity for the back end is primarily increased the tester capacity by adding more testers and also getting additional manufacturing partners qualified to mitigate the potential geopolitical risk. On the front end, we have been working with a leading partner in increasing capacity we are mobilizing all different ways in increasing the prepayment, the CapEx or some other means. But I think though are the same to bring additional capacity to support the growth.

As for your question about NPL, yes, absolutely. That's a strong trend. The primary driver is to increase the bandwidth density. As data -- total capacity increased dramatically, while shoreline space is limited, they need to have the density, high-density packaging. We're going to be benefiting from that, and we're currently engaging, I don't know, 10, 15 different programs with all the module manufacturers and some of them were directly tied to end customers. So net-net, we're going to be benefiting from that.

We are already a leading provider of TIA arrays and on laser arrays, especially linear right version of excellent as well. So I just talked about our initiative to start photodiodes and by co-optimization between TIAs and photodiode bring to our customers even better solutions. So it's a great opportunity for us. That can be translated into a new growth driver for us in the future.

Our next question is from Rick Schafer with Oppenheimer & Company.

I'll add my congratulations to you guys. Great quarter and need a better outlook. If I could, I'll just start a quick one on LoRa. I mean the run rate there was fairly $150 million just a year ago. I mean, we heard your guide on that home. I mean that's close to $60 million a quarter now so well over the 150 just in the last 12 months. So is 20% still a right bogey because I think you're going to be doing about 3x that growth in the third quarter.

We certainly matched the 20% ceiling with Q3, just the -- we're seeing sequential growth of 15% year-over-year will be translating into 65%. So that is certainly higher than 20%. And we benefited from now 3 pillars of growth not just the traditional LoRaWAN in supporting the industrial and commercial applications, but also LoRaPLUS in security and smart home, smart buildings. And now with Amazon and in the Sidewalk and in strong engagement and their plan to deploy internationally start from North America, expanding into Europe and Australia. We see that is going to be a strong growth driver as well. So I do expect year-over-year, [indiscernible] going to be better than 20% going forward and is sustainable.

And if I could, I wanted to just get a little bit more color on FIFO. Obviously, you're investing in capacity there. I think you've talked about tripling that capacity by the end of the year. I don't know if you could level set us on where we are in the process, if there's any symptom of a design funnel or revenue formal or anything you could share on that? And then as part of your answer, I'd be curious, I mean, folks are talking about CW laser channel densities really rising, right, going up. So I'm curious how much does that pull the need or create the need for higher density drivers and TIAs? And then if so, what does that do to the complexity and the barriers to entry there for your competition?

Yes. So first, we'll start with the HFO acquisition. Certainly, we have been the proud owner of that asset for the last 5, 6 months. We have made tremendous progress in upgrade the line and also getting more wafer start reaching out to the customers and with Semtech behind the asset, the customer confidence level has improved dramatically. So we're not only with the 3 anchor customers increasing their -- they are increasing the demand, but we are able to expand into other key customers on the game chips.

Now we have been, as I mentioned in the prepared remarks, sending high-power lasers to 5, 6 module manufacturers they've been evaluating and really satisfied, really very excited about our best power conversion efficiency in the bean performance and over-temperature performance. Those are pretty ideal in having one CW laser split into 4 channels or 8 channels for high bandwidth transceivers, like 1.60 and 3.2T. We also have the product, we start sampling to customers and semiconductor optical amplifiers.

That's almost like a game chip, you get the tempos currency, you get the amplification. So that is the foundation we are using. And the capacity is limited, as I mentioned, we are going to be increasing capacity by bringing more testers in the back end first. then for the fab capacity, we're just forced it to be able to acquire an already fully facilitize the fab in a close proximity to the current facility. So that allow us to increase the fab capacity by 3x to 4x by the end of the year. So we're on track for that. As for MPO, the high density, certainly, when you do the high density, the spacing between different elements is becomes smaller when you go high speed at the cross top and all the other performance the packaging need is different.

It's representing another new set of challenges, that's why we expand not only from the fiber edge to photonics that will allow us to do co-optimization to improve signal integrity. And definitely, the industry is welcoming our move, and we have increasing engagement with the module manufacturers and hyperscalers because of that expanded capability.

Thank you, Rick. Our next question is from Sean O'Loughlin with TD Cowen.

I'll add my congratulations on, obviously, some really strong momentum across the business here. I wanted to start with just a high-level question on data center strength. Really strong outlook in the forward quarter and talking about acceleration through the back half. But I think in your prepared remarks, both Hong and Mark, you both mentioned the copper edge in a high-volume ramp. I think that aligns with some of your past comments, but maybe you could just talk about how to think about the contributors to growth and how that aligns with some of your commentary on TIA share in -- towards the end of this year on 1.6?

Yes. Thank you, Sean, and that's a good question. So maybe I will use this opportunity to just review the progress we have made in expanding our portfolio in the data center play. So we certainly -- the reason the investment community over indexing on copper edge is because that's the first time I think the investment community paid attention to Semtech 2 years ago, we developed this redriver, our leader equalizer solution, which that can be embedded in ACC cable to interconnect to adjacent regs. And so that continues to be a really de facto standard for the industry for 1.6, 2.4, 3.20 and going forward. Copper scale-up continues to gain momentum, especially linear equalizer on board.

So we got multiple engagement and some of them will reach the finish line in the near term. [indiscernible] the FiberEdge 2 years ago, 800 gig, we had the market share about 18%. So over the 2 years, we have grown the market share well over 50% for 800 gig. 1.60 is just inflecting. So 1.6%, as I said in the prepared remarks, we are expecting to exit this fiscal year by January with better than 50% of market share. So now you see the FiberEdge area, not only we are gaining shares. The volume has increased dramatically for 800 gig, for example, transceivers from 2 years ago with 20 million units a year to this year, probably 90 million units a year. Regaining shares, we benefit from increased volume and we are expanding the product offerings 800 gig and 160 and drivers.

The driver revenue is to come -- we got a wonderful product in the evaluation will be contributing to the revenue very meaningfully. A few months ago, we acquired [ HiFo ] and mark the beginning of our journey into the photonic area. And we're going to be expanding and having meaningful play in that area as well. So now I'd like to encourage everyone to look at the data center play for Semtech is not just a top edge. Copper at will definitely be a significant part of the data center revenue. But think about the fiber Edge, the leading share of the TIA and drivers and photonics offering from game chip to lasers, to photodiode to SOA. So we're going to be continuing to expanding our portfolio to become a key player in this area.

Great. And just if I could ask a follow-up and get Mark into the party here. The gross margin expansion quarter-over-quarter is striking even if you're just looking at the consolidated and not isolating the held for sale business. Just wondering, I guess, a question on how much of that can be thought of as mix? And if data center continues to stay at this percent of revenue, is that something we should expect levels that we should expect to continue? Or is there some onetime thing? And then maybe as part of that, just talk about the capacity expansion. And you've heard some of your suppliers talk about what they're seeing on the pricing side? And what's giving you confidence on the margin sustainability?

Yes, Sean, I can address that and try to address that. So the sale is cellular module business. I expect will result in over 500 basis points of gross margin improvement. That's the structural change. That's a significant drop change that we see in our gross margin profile. And I provided some details in my prepared remarks, but I think it's helpful to walk through those figures again.

So from Q1 to Q2, our consolidated adjusted gross margins increased from 53% to 54.5%. That's 150 basis points, largely on mix. Q2 adjusted gross margin, excluding modules, was 59.7%, which is an incremental 520 basis point increase. Then we move to our Q3 guide. Our consolidated adjusted gross margins are projected to increase from 54.5% to 58.3%, 380 basis point increase. And then on top of that, we add 560 basis points to arrive in adjusted gross margin guide, excluding the models at 63.9%. So you have the 500 basis point plus gross margin improvement just based on the structural change.

But the 150 basis points to 380 basis points, that's mix. I think a good starting point post divestiture, is that 64% gross margin. And as you're seeing the mix change. I mean mix is quite a powerful broader for Semtech as 1.60 continues to inflect as LoRa continues its strong growth and 800 gig, maybe just to briefly address pricing, right? We're not really seeing price erosion to 800 gig. That all compounds into some very strong mix changes, which is that 150 to 380 basis points that we demonstrated in Q2 and Q3.

Our next question is from Christopher Rolland with Susquehanna.

Thanks for the question, guys. So this was kind of asked, but maybe more simply, the data center guide or next quarter's guide driven by data center. What exactly are -- like what did you not anticipate that is driving this? Is it the cycle? Is it LPO? Is it really that TIA attach that you're talking about? Or is it copper edge? What kind of drove the marginal upside versus perhaps your expectations or even the Street's expectation guys like me.

Yes. Chris, that's a good question. So if you look at the data center portfolio, we know 800 gig is going very well. We've got a line share, and we continue the volume increase. We also know the copper edge 1.60 timing that has been largely on track and going with the schedule. If you say upside came from a little bit earlier inflection, for 1.6 fiber edge. We know we are intense engagement with our module manufacturers, as I said, and their customers. So we were just not very sure about the qualification timing, and that's why we're a little conservative in guiding for Q2 at the time.

Now we have all the backlog and the customers want parts tomorrow. So we definitely have a very high confidence and conviction for Q3 and Q4. So if you say what's different from a few months ago, I mean that's just the qualification timing. When customers need a solution they go out of the way they accelerate the pace of new technology adoption. So that is -- I've seen that before. But this is really in a way it's unprecedented from the hyperscalers to module manufacturers to the technology providers, component providers, working all together to accelerate that pace. Does that make sense?

Yes, that may -- that totally makes sense. And then perhaps a follow-up, just as you ramp FOFOs me, and you have all these new products coming into this portfolio, and you talked about getting to high double-digit per transceiver content for you guys -- can you walk us through just a time frame of when you expect these products to ship in volume to the market whether it's these high-power CW modules, photodiodes, SOAs or anything else that, that acquisition will be able to provide?

Yes. So Chris, we only get into this area, as I said, for 5 months or so. We certainly have a great plan and great ambition. Right now, the ongoing product shipping in volume is a gating chip, and we're going to be having the CW, high-power CW lasers and SLA available for sampling and qualification from the customer side in a couple of months. And -- but the significant increase in content in optical transceivers as we said before, it's more like 3.2. Because we see the ramp of the fiber edge from 1.60, that means the customers are already wrapping up the qualification and getting ready for volume production. If they don't have a solution now, they'll probably be late.

We wanted to catch the next wave. So that's a 3.2T.And good old high-power CW laser work still is the most needed for that application. By then, we wanted to make a photo diode available as well because when the data rate is going higher than 200 gig, and need every bit of help from electronic component and photonic component so the co-optimization allows us to provide a cross-reference design solution to customers. That is also very much needed for 3.2T. So to answer your question, really the significant content increase in one optical transceivers will be coincided with a 3.2 transceiver cut-in.

Our next question is from Harsh Kumar with BMO Capital Market.

Yes. Hong, Mark and Mitch, congratulations on stellar quarter and Stellar guide. I had a one multipart and then another follow-up. Hung, you talked about 3.2 being the catalyst for your products catching growth. Could you talk about what the timing for 3.2 is as you see it in the field? And then I want to push back on your commentary a little bit as well. You talked about your content going from high single to kind of high double digits. But when I look at all that you have in the pipeline, photodetectors and game chips and drivers, et cetera, I would think the content would be more than teens. Are you just being somewhat cautious here? Or is there any other reason for that commentary?

As, first of all, thank you for initiating the coverage, and we look forward to working with our new platform. So probably I confused you the high double-digit means 80, 90 instead of 18, 19. So that's the content we're talking about.

No, no. Thank you for that clarification. I appreciate it. And 3.2 timing, Hong?

3.2 timing, I would say, probably in 18 months or so, but I think the design window will start opening up at about in 12-month period of time. So then the early movers, they probably be 18 months from now. And -- but I think the meaningful deployment will start probably in 2 years. even 800 gig will have a really very healthy runway over the next 2 years.

And then for my follow-up, if I can ask you about ACC and LPO? The reason why I'm asking you, you're coming out as the clear leader in those two technologies. You talked about it, I think, a little bit more positively in this call. Can you help us still give an idea of what we should expect the growth rate to be, let's say, exiting this year or at some point in time next year, what can these 2 businesses do?

Yes. So the ACC, we definitely have the clear visibility with the leading hyperscaler. We are going to be having the volume deployment start from. But right now, all the cable manufacturers are ordering and increasing quarter-over-quarter. But the inflection is going to be started from the -- then in the meantime, we see so many design activities of linear equalizer on board. The dynamics, we start understanding this emerging market better now. ACC adoption is more coincide with the new platform design.

So they wouldn't be yanking out the AEC currently in use and to put in ACC. But linear equalizers board design is happening on the Board level. So we get a lot of activities. We continue to be very bullish on that market. As for PO, it's almost -- we had a meaningful revenue from Q1, and that has been increasing moderately. But that deployment really get the industry give them the confidence of the linear architecture, and it works really well. So that evolved into NPO in some form then even the XPO is including the LPO form with the linear equalizer -- well, the linear architecture instead of realtime. So I think in the future, the LPO impact not only as a stand-alone transceiver, but also the proof of the concept approval of the technology get incorporated in more integrated form factor like NPL

Our next question is from Joe Moore with Morgan Stanley

Congratulations. Can you talk about the strength in 800 gig. You talked about that persisting for a while. What's your visibility into that? I know 1.60 is the big ramp up. A seems quite strong. Can you talk about that dynamic a little bit?

Yes. So Joe, we enter into the year for 800 gig, the industry is forecasting 50 million transceiver units to be consumed. Now we are here in the number at 90 million and we have a very healthy backlog for FiberEdge to support 800 gig. And that is continued and we're just getting -- picking to our new booking report this morning. So existing customers, they are increasing the demand now decreasing. In the meantime, the 1.60, it's just starting. And so the Q3 will be the first quarter for us to really have a pretty significant revenue, as I said, between 1.6 fiber edge and 1.6 copper edge, we will have -- they will be surpassing 50% of total data center revenue. So 1.6 is gaining a lot of momentum and gaining momentum fast.

That's very helpful. And then can you just discuss like-for-like pricing? Are you seeing any changes really in any part of your business that just particularly on the optical side, any change in pricing there?

Yes. So pricing, it's very favorable. Typically, as I said before, you offer 2x performance, you may be able to get 10% premium in the beginning, but it settled at a lower level. But right now, availability is more important to the customers and then the pricing is almost. There's no erosion to be expected in the near term and at least not for any orders we booked in the backlog. Our cost is increasing slightly, and we are able to, in most of the cases, work with the customers and pass along the cost to them.

And -- but we are here to build a long-term relationship with our customers, we are very mindful and not being viewed as using the seller advantage to gouge our customers. So we're working with them in a partnership fashion, but we are able to pass along the cost increases. So that's why when Mark talked about the gross margin we're expecting the trend to continue to grow because of favorable product mix in the new product, we have a higher gross margin.

Your next question is from Tore Svanberg with Stifel.

Yes. And congratulations on the record quarter. Hong, so you're going to be at $0.5 billion run rate in data center next quarter. I'm just curious as we sort of think about $1 billion in data center revenue, how should we think about the mix between FiberEdge, CopperEdge and all the new products that are coming online?

So that's probably the one we're going to be providing more detail in our October Investor Day because we plan to get the different -- the TAM for the different applications and our market share and so that you can have a more comprehensive view for multiyear model I hope you can come to that event, sorry.

Sounds good. And as a follow-up, a similar question for Mark. So ex modem will be at 64% gross margin. How should we think about the margin contribution from some of the newer products like PB, CW lasers and so on and so forth. Are they going to be at that corporate average or perhaps even above?

It should be above. So all the areas, it's that you just mentioned, CW lasers, especially. They're the data center gross margin, which is accretive to the corporate gross margin average.

Our next question is from Craig Ellis with B. Riley Securities.

Yes. Congratulations on the stellar performance, and thanks for sneaking me in guys. I wanted to look at the business through the 1.60 lens. So this sort of follows up on part of what you got to with Joe. But can you clarify what you're looking for, as things get going in the third quarter as a percent of mix. And then, Hong, it sounds like we're starting stronger in fiber edge and copper comes along. So how would mix evolve for 1.6 as a percent of total as we look out to next year?

So Craig, thank you for the question. In Q4 -- Q3, the 1.60 is already surpassing 50%. I can just only imagine that it's going to be continue to grow north of 50%. But the 800 gig continued to be very strong, and we have the Tri-Edge legacy products continue to kicking very strong. So but the trend is going to be growing the percentage of 1.6 is going to be higher and higher.

Got it. And then I wanted to go back to your comments where you indicated that beyond the near-term 45% Q-on-Q growth for data center in the third quarter and the 10% year-on-year growth, we could see acceleration. And the comment on backlog just suggests that you've got tremendous visibility out into fiscal '28. Can you comment on where that would be relatively greater. And what some of the interactions are like with customers that are booking out that far? Is it really just you becoming a lot more strategic to the road map? Or is it that supply sufficiency point? Just help us see what you're seeing.

Yes. Thank you, Craig. I think it's all of above. You see our product performance is great, and we can provide availability and also we are providing excellent services. So that has been the key drivers for us to gain shares.

Our next question is from Cody Acree with the Benchmark Company.

And congrats on the progress. Hong, maybe just a follow-up on the last question. With the bookings and backlog accelerating here in the second half. Any quick thoughts on what kind of how long into '28, does that backlog extend? And any thoughts on what kind of growth that might support next year? SP1 So the backlog for the remaining of this fiscal year, I would say, for our targets, it's our booked. And for the next year, we probably over 70% there. And -- but the momentum is so strong. So we are going out to get more capacity secured. And so far, I would say the visibility side for the next fiscal year, we feel very confident about it. Do you think Hong that you have upside to -- with that capacity addition effort. Do you think there's room in the industry for you to secure more and to continue to service upside?

Yes, between 50% and 100%, there's room for that. And we definitely wanted to be able to provide our customers for the product they need.

And then lastly, you've had a lot of success with your lead hyperscalers for ACC. Any thoughts on the expansion beyond that lead hyperscaler? How broadly do you think this gets adopted? And maybe what are the some of the challenges left to full adoption? Is it interoperability, support or cable qualifications? So they -- yes, they -- at this point for them, the availability, the interopbility, probably more important for them. We have not seen many activities in driving us to interact with other industry participants. Any thoughts on breadth of adoption?

So is it going to be more -- what time is going to be broader, and we'll start with the leading one. And there are multiple engagement. It's an ongoing, some of them is going to be reaching to the finish line. So I guess we're at the time, right? But I do encourage I know that we still have some in the queue, but come to our October 15 investors event. So we definitely wanted to provide very comprehensive information on our technology road map, differentiation, TAM, our share and multiyear model, so that will help you to build a multiyear financial model for us.

Our last question is from Scott Searle with ROTH Capital Partners

Thanks for sneaking me in. Congrats on the quarter and incredible outlook in terms of data center and lower Data center has been covered pretty thoroughly. So maybe hopping over to lower for a second. Just in terms of -- could you calibrate us quickly, you've been moving away from the China mix. It had been down under 50%. I want to just you have a better idea about how that was progressing in the July quarter. And looking at the growth that you've seen from the first quarter to guidance now in the third quarter, it's up 50%. How big is Amazon now factoring into that? Are they over a 10% customer? And kind of stack ranking the guidance into the third quarter. Is that mostly Amazon? Is it lower plus? Is it something else that what's really driving the outlook?

And lastly, to follow up now on the IoT side of the equation with modules now on the path to be divested. Other elements have periodically been core and then noncore on the router gateway on the IoT platform front. I'm wondering what the current thoughts are in terms of their continued inclusion as a core portion of Semtech going forward?

Thank you, Scott. On LoRa and the majority of the revenue is still lower on an LoRa start kicking in, probably representing about 20% to 25% of total revenue mix. and site work is still, at this point, is nominal, and this year will probably be high single digit, but to get a lot of potential once we are able to piggy back into the consumer at Amazon's scale. As for the portfolio optimization, as we mentioned, it's a continued journey. And so far, we like the portfolio we have after the divestiture, but we'll continue to evaluate addition or optimization effort. But we are focused on getting the current deal to the finish line, and so which lead to the closing of the sale of the cellular module business.

There are no further questions at this time. I'd like to hand the floor back over to Mitch Haws for any closing comments.

Thanks, Paul. That concludes today's call. We look forward to seeing you at various investor events during the quarter, including our Analyst Day on October 15. With that, good afternoon, everyone.

This concludes today's conference.

Good day, and thank you for standing by. Welcome to Semtech Corporation's First Quarter 2027 Earnings Conference Call. [Operator Instructions] Please be advised that today's conference call is being recorded. I would now like to hand the conference over to Mitch Haws, Senior Vice President of Investor Relations for Semtech. Please go ahead.

Thank you, and welcome to Semtech's First Quarter 2027 Financial Results Conference Call. Participants on today's conference call are Hong Hou, President and Chief Executive Officer; and Mark Lin, Executive Vice President and Chief Financial Officer. Today after the market close, we released our unaudited results for the first quarter ended April 26, 2026, which are posted along with an earnings call presentation to our Investor Relations website at investors.semtech.com. Today's call will include various remarks about future expectations, plans and prospects, which comprise forward-looking statements. Please refer to today's press release and see Slide 2 of the earnings presentation as well as the Risk Factors section of our most recent annual report on Form 10-K for a number of risk factors that could cause our actual results and events to differ materially from those anticipated or projected on today's call. You should consider these risk factors in conjunction with our forward-looking statements.

We will refer primarily to non-GAAP financial measures during today's call, and we'll also be referring to results for our first quarter of fiscal year 2027, unless otherwise noted. Please see today's press release and Slides 3 and 4 of the earnings presentation for important information regarding notes on our non-GAAP financial presentation. The press release and earnings presentation also include reconciliations of our GAAP and non-GAAP financial measures. With that, I will turn the call over to Hong.

Thank you, Mitch. Good afternoon to all of you joining today. Semtech is off to an exceptional start in fiscal year 2027, delivering record quarterly revenue supported by very strong bookings and backlog. We drove strong sequential and year-over-year revenue and earnings growth, expanded our data center and LoRa design win pipeline, all while advancing our R&D and strategic initiatives. We believe we have built a robust foundation to solidify and expand our presence in key markets. My strong conviction in Semtech's positioning is rooted in the transformation we have seen across Semtech employees, motivation to engage and the partner across the ecosystem and an appreciation for the benefits of collaboration.

The time I have invested has been energizing, and I have appreciated opportunities to join my Semtech colleagues in meeting with hyperscalers, device designers, end customers, module manufacturers and our technical partners to understand their technology roadmap firsthand. Those conversations shaped our R&D priorities and give us first insight into where the industry is heading and how Semtech can remain at the forefront. My team and I spent time with the key suppliers and distributors to round out our understanding of how Semtech can partner with our customers to win from design to delivery.

Looking at Q1, revenue was $291 million, up 6% sequentially and up 16% year-over-year, driven by continued outperformance in both data center and LoRa. Adjusted diluted earnings per share were $0.51, up 34% year-over-year. In addition to delivering strong revenue and earnings growth, we are laser-focused on executing our portfolio optimization initiatives. We are pleased to report the divestiture process for our cellular module business is at the final stages. Discussions which are transition and integration in nature are progressing well. We remain confident this business is a compelling opportunity for the right acquirer, and we look forward to bringing this process and transaction to a successful close.

Now let me move on to a discussion of our end markets. For Q1, infrastructure net sales were $98.8 million, up 14% sequentially, up 36% year-over-year, strongly supported by our growing data center business. Our net sales for data centers in Q1 were a record $71.6 million, up 14% sequentially and up 39% year-over-year, benefiting from strong demand across our broad portfolio, the result of sustainably increased customer engagement, portfolio alignment and supply assurance. The strength is anchored by our strong position in our 800-gig FiberEdge portfolio. Demand for our leading TIA solutions is exceptionally strong growing across a wide range of transceiver programs. Based on our differentiated technology and ability to supply both established and emerging module suppliers have qualified us on several new sockets, some on a sole-sourced basis. We understand the module suppliers are winning shares in key mega data center deployments.

On 800-gig linear pluggable optics or LPO, our FiberEdge linear TIA and driver solutions are deployed by several leading hyperscalers across both the U.S. and in China, which contributed to sequential LPO revenue growth, a trend we expect to accelerate over time. We remain confident our foundation in 800-gig will continue to drive revenue growth throughout this year, further augmented by significant opportunities at the 1.6T shipments launching in Q2 and gaining momentum in the second half of the year. On 1.6T optical, we generated significant design wins with the major optical module makers for their 1.6T transceivers incorporating the latest generation DSPs. This contributed to exceptionally strong bookings and backlog to support module ramps in the second half of the year.

We are also seeing increased conviction from hyperscalers around 1.6T linear receive optics or LRO and LPO as a preferred solution for first layer scale-out fabric due to the substantial power savings. Looking further ahead, we are participating in the development of the NPO or near package optics MSA and to see NPO as a meaningful content expansion opportunity for Semtech as 800-gig and 1.6T successes in LPO and LRO give hyperscalers confidence in the next evolution of high-density and low-power optical solutions. We're also developing derivative components with the same core IP in different form factors to support several NPO projects for leading hyperscalers.

We are actively participating in and support XPO MSA and many XPO module designs incorporate our FiberEdge chips as they do in the OSFP modules. We believe XPO provides a very compelling alternative to CPO scale-out. By leveraging liquid cooled capabilities, proven technologies and components and established innovative optical module ecosystem, XPO can provide significant rack space savings along with improved serviceability and better reliability.

On the copper side, we are very enthusiastic on CopperEdge deployment. ACC continues to gain meaningful traction. Customers evaluating ACC against incumbent solutions are seeing compelling advantages in link margin versus direct attached and power savings versus DSP-based solutions. Consistent with our expectations, in Q1, we started shipping CopperEdge 1.6T ICs to our cable partners for deployment at a U.S. hyperscaler. In onboard integration applications, including active backplane, CopperEdge linear equalizers are gaining momentum. Just as we were confident of ACC's acceptance and ramp in the market, we have increased confidence this engagement will convert into design wins and widespread market adoption. Based on our engagements across different sectors of the industry, we believe we are creating a multiyear pipeline of CopperEdge opportunities, design wins and revenue.

Looking forward, we are excited by the opportunity from the HieFo acquisition we completed in March. HieFo is reported in our Signal Integrity Products segment and its indium phosphide photonic products are reported in the data center end market. These products are a strategic building block in 1.6T and 3.2T optical modules and a key pillar in our strategy to support next-generation data center requirements. We believe our Gain chips have become the industry standard, providing higher power and serving as reliable building blocks in tunable lasers for coherent modulation applications in metro and data center interconnects. Gain chip demand currently exceeded our supply, but our capacity expansion plan is on schedule.

We believe our continuous wave of CW Laser design is uniquely differentiated to deliver higher conversion efficiency, superb far-field beam profile and over temperature performance and narrower linewidth. These lasers have been sampled to and evaluated by several major module manufacturers for coherent light modules in scale across applications. Concurrently, we are optimizing our laser drivers and TIAs for coherent light applications. We plan to provide a comprehensive suite of photonic and electronic component solutions for these emerging high-volume applications.

In addition, we are also working with key customers to make dense wavelength division multiplexing or DWDM lasers optimized for emerging CPO scale-up applications based on the newly established OCI MSA. This is exactly the kind of strategic investment we believe creates durable and compounded value. Not just a single product win, but a platform capability that strengthens our position across the broad spectrum of optical architectures our customers are building to work. Semtech is uniquely positioned at this intersection with a portfolio that [ spans ] scale up, scale out and scale across, addressing the full hyperscale interconnect stack across both near-term deployments and next-generation architectures at 800-gig, 1.6T, 3.2T and beyond.

Finally, given the strength and the depth of our backlog, expanding design win momentum and the 1.6T FiberEdge and CopperEdge inflection building into the second half, we are targeting 35% sequential revenue growth in Q2 for data center. which would represent 85% growth over the same period last year. Based on the current order trend, we expect accelerating demand throughout fiscal year 2027 and beyond.

Now moving to the high-end consumer end market. Net sales for Q1 were $38.4 million, up 5% sequentially and up 8% year-over-year. Our TVS business continues to demonstrate impressive resilience and momentum with revenue growth outpacing underlying handset volumes. We continue winning shares and expanding content at the premium brand handset manufacturers. Our differentiated technology is aligned with the right customers and the alignment is translating into consistent design win momentum that we expect to continue.

Beyond handsets, we are actively expanding the TVS franchise into higher-value applications. Our newest SurgeSwitch solution is the industry's first circuit protection device to deliver near constant clamping voltage for high-voltage power delivery applications, addressing a meaningful protection gap as more demanding power standards extend into rugged mobile devices and high-performance portable systems. These are environments that require consistent, reliable protection across extreme temperature range and operating conditions, and our solution is purpose-built to meet that bar. We see this as a natural and incremental content expansion that broadens the TVS opportunity beyond our core handset market.

We continue to expand our PerSe capacitive sensor design wins in specific absorption rate and smart wearable applications. The addition of the force sensor business enriches our high-end consumer portfolio, expand application verticals and pull through some CAP and TVS sales with the same customer base. The synergies we have played out as we planned. For the high-end consumer end market, we expect sequential revenue growth driven by improving seasonality layered on top of the share and content gains that are becoming a defining characteristic of this business.

Moving to our industrial end market. Q1 industrial net sales were $153.9 million, up 2% sequentially and up 8% year-over-year, driven by another great quarter for LoRa. LoRa-enabled net sales were $44.5 million, up 12% quarter-over-quarter and up 14% year-over-year, supported by continued expansion across several application verticals such as smart utilities, smart building, smart city and asset management. As Edge AI transitions from concept to deployment reality, LoRa Plus is emerging as a key enabler. Our fourth-generation LoRa platform delivers dual band capability while dramatically expanding data throughput to 2.6 megabit per second, a step change increase that unlocks new AI application classes.

At the same time, LoRa Plus maintains the best-in-class sensitivity, multiprotocol flexibility and ultra-low power consumption that defines the LoRa advantage, preserving the extended reach and the long battery life our customers depend on. We are seeing LoRa Plus gaining traction across a broadened set of use cases. LoRa connected public safety sensors can now transmit high-fidelity audio and AI verification rather than simple alert. In health care, fall detection systems can relay visual confirmation before dispatching responders. In the industrial environment, predictive maintenance sensors can analyze vibration, thermal and acoustic profile with a level of detail that legacy low-power sensors could not support.

We have established 3 distinct and complementary pillars of our low-power connectivity platforms, LoRaWAN for industrial and commercial deployments, LoRa Plus with multiple protocol flexibility for smart home and security market and Amazon Sidewalk for mass market consumer applications. Together, these growth vectors give rise to an accelerated growth in our LoRa business as we target LoRa revenue at an all-time high with greater than 15% sequential quarterly revenue growth for Q2. Our IoT Systems and Connectivity business recorded Q1 net sales of $88.3 million, down 2% sequentially and up 2% year-over-year.

Our newly released AirLink RX400 and EX400 routers are generating strong industry reception. These are industry-leading low-power 5G cellular system purpose-built for mission-critical applications and the feedback from customers has been consistently positive. I recently attended our annual AirLink Partner Summit alongside national carriers, integration partners and value-added resellers and the enthusiasm for both the router performance and our upgraded AirLink management software was clear. The close collaboration with our channel partners position us to scale successful use cases from regional to national deployments and accelerate this high-margin business.

We are off to a strong start and the momentum is building. Our data center business is firing on all cylinders. LoRa is entering a new chapter of growth and the strategic decisions we have made in prioritizing key R&D efforts, enhancing supply assurance and portfolio optimization are all translating into tangible results and the financial flexibility to pursue strategic opportunities. Our priorities for fiscal 2027 are straightforward. First, accelerating growth by supporting customer ramps with availability and operational excellence required to compete in this capacity-constrained environment.

Second, intensifying R&D investment to add new growth drivers and deepen our solution differentiation, specifically in component offerings for coherent light, CPO, LoRa and sensors; and third, continuing to transform Semtech by strengthening our culture and completing the initial steps of portfolio optimization. We are just getting started and the opportunities ahead have never been more compelling. With that, I will turn the call over to Mark for additional details on our financial results and our second quarter outlook. Mark?

Thank you, Hong. For Q1, we recorded a ninth consecutive quarter of net sales growth with record net sales of $291 million, above the high end of our outlook range. Net sales grew 16% year-over-year, while adjusted diluted earnings per share of $0.51 increased 34% year-over-year. Net sales trends by end market, reportable segment and geographic region are included in the accompanying earnings presentation. Adjusted gross margin was 53%, 20 basis points above the midpoint of our outlook and total semiconductor products gross margin was 60.7%, 30 basis points above the midpoint of our outlook, both reflective of favorable mix from our data center and LoRa portfolio. For our Signal Integrity Products segment, Q1 gross margin was 62.7% compared to 67.4% in Q4.

Q1 is the first quarter of operating our recently acquired indium phosphide facility. This facility is in ramp mode to meet very strong customer demand, and I'm pleased with the integration team's progress in meeting its operating and financial targets. We continue to expect gross margin contributions from our 1.6T data center portfolio to be accretive to both our total semiconductor products and Signal Integrity products gross margin. Gross margin for IoT Systems and Connectivity was 35.8%, up sequentially from Q4's 31.6%. Adjusted net operating expenses were $95.1 million, slightly favorable to the low end of our guidance range, reflective of timing on project-related expenses.

Demonstrating the operating leverage in our business, a number of metrics were favorable to the high end of our guidance range, including adjusted operating income of $59.3 million, adjusted operating margin of 20.4%, adjusted EBITDA of $66.4 million and adjusted EBITDA margin of 22.8%. Reflective of capital structure changes, Semtech remained in a net interest income position in Q1. We recorded adjusted diluted earnings per share of $0.51, above the high end of our guidance range. Operating cash flow for Q1 was $36.2 million, sequentially down 41% from $61.5 million and up 30% from $27.8 million a year ago.

Free cash flow for Q1 was $28 million, sequentially down 53% from $59.1 million and up 7% from $26.2 million a year ago. Q1 operating and free cash flow reflect fiscal year 2026 annual bonus payments. In addition, net acquisition consideration of $29.2 million is reflected in our Q1 ending cash and cash equivalents balance of $163.3 million. Principal amount of debt was $503 million, unchanged from last quarter.

Now turning to our outlook for the second quarter of fiscal year 2027. We currently expect net sales of $328 million, plus or minus $5 million, up 13% sequentially and up 27% year-over-year at the midpoint, with growth expected across each of our 3 segments. We expect net sales from our infrastructure end market to increase sequentially with projected sequential data center growth of 35%, supported by accelerating shipments of 800-gig and 1.6T components. We expect net sales from our high-end consumer end market to increase, benefiting from improved seasonal trends, market share gain on our TVS products and contributions from our sensing portfolio.

We expect net sales from our industrial end market to broadly grow with accelerating contributions from LoRa, IoT Systems and Connectivity and Industrial TVS. Based on expected product mix and net sales levels, we expect adjusted gross margin to be 54%, plus or minus 50 basis points. Midpoint, this equates to an increase of 100 basis points sequentially and 80 basis points year-over-year. Our gross margin outlook for our total semiconductor products is expected to be 62.1%, plus or minus 50 basis points. At the midpoint, this equates to an increase of 140 basis points sequentially and year-over-year, reflective of stronger data center and LoRa mix. Adjusted net operating expenses are expected to be $105.2 million, plus or minus $2 million.

Included in this outlook is increased R&D spend to accelerate time to market on key data center projects, along with SG&A that declines as a percentage of revenue. We have demonstrated strong returns on our R&D investment and believe we remain prudent on SG&A spend. This results in adjusted operating margin at the midpoint of 21.9%, up 150 basis points sequentially and up 310 basis points year-over-year. Adjusted EBITDA is expected to be $79.2 million, plus or minus $2.3 million, resulting in adjusted EBITDA margin at the midpoint of 24.2%, up 140 basis points sequentially and up 230 basis points year-over-year. We expect adjusted interest and other expenses net to be approximately $0.5 million. We expect an adjusted normalized income tax rate of 17%, consistent with last quarter's outlook.

These amounts are expected to result in adjusted diluted earnings per share of $0.61, plus or minus $0.02, up 20% sequentially and up 49% year-over-year at the midpoint based on a weighted average share count of 97.7 million shares.

Thank you, Mark. We can now turn the call back over to the operator for the question-and-answer session.

[Operator Instructions] Our first question comes from the line of Rick Schafer with Oppenheimer & Company.

Congrats on another good quarter. If I could. Maybe my first question, I didn't hear you mention it on the call, and I'm sorry if I missed it, but where do we sit now with ACC MSA specs? And when do you expect that we'll get those specs finalized? Because I'm really curious if you think interoperability is holding back the ACC ramp at all? And if you think ratification, I mean, does that create sort of that dam burst moment with some of the other CSPs? So that's my first question. I'm just curious what you think about that.

Yes, Rick, thank you. That's a good question. I did not mention in my prepared script, but that's a good point. So MSA, they announced right around the DesignCon time. They're still working on finalizing the specification. You are right. When every industry participants, they are on the same page, that will help to accelerate the adoption of the ACC. Currently, they are working on the common denominators, as you can imagine, from different cable manufacturers. There are some related to the cable design. There's some performance related to the manufacturing processes. They just wanted to segregate those different impacts and define a standard and specification that every cable manufacturer can go with and sign off for. So yes, that can be a catalyst for more accelerated adoption of ACC.

And for my follow-up, just on the supply side, I mean, you guys are clearly seeing a step-up in growth in 2Q. And I heard you loud and clear talk about acceleration through the year on the top line. So my big question is, I mean, do you see any curves on your ability to capture some of that upside that's coming through, capture some -- continue to capture share like you're doing? I mean -- and maybe if you could level set us on what top line -- basically what top line currently, your current capacity actually supports?

Yes, Rick, that's a great question. This is a time capacity is king. So we anticipated that we started working on the capacity and availability. You hear me like a broken record to talk about making it available over the past quarters. And we started about 18 months ago. I am so happy to say that in this very supply-constrained environment, we're doing quite well. We have the availability. We have the right product. We have the right customer engagement. We can support even the drop in orders. That's why our momentum is building so fast. Going forward, looking into the visibility, the booking and the outlook opportunities, clearly, the capacity we have put in place is not going to be serving us well going forward. And we have already started effort lining up with our foundry and OSAT partners, adding capacity in testing. So we are building capacity further double or triple the current capacity. And that's how our conviction is supporting our current planning process.

Our next question comes from the line of Sean O'Loughlin with Cowen and Company.

Congrats on, obviously, the really solid results and guidance. Hong, one of the things that I perked up at during your prepared remarks was you mentioned the narrow-linewidth CW lasers in the context of coherent light, but then you talked about DWDM and the OCI MSA in the CPO scale-up side. And I thought that was interesting because I think that the CW and ultra-high power lasers is kind of how we're going about CPO for scale-up today, but maybe you're anticipating a shift in how we're going after that in the future. And I was just wondering if you could expand on that and talk about sort of where the HieFo business is investing their time and where there -- is this an organic investment that's coming out of the asset that you acquired? Or is there still more to go on either the hiring front or the CapEx front to make this a reality?

Sean, thank you. That was a very good question. Well, good questions. First, the coherent light clearly is going to be a major application. As data center right now consume a lot of power, you cannot build this mega data center in isolation in more a cluster of data centers. So scale across to be able to interconnect with high bandwidth the different data centers is very, very important applications. I mentioned about my customer sensing trips and talk to 10 different leading module manufacturers. They all supporting our conviction and the vision of the coherent light is going to be ramping up in the mid-2028 in production. We, through the acquisition of HieFo, have the right product. Our DFB laser CW lasers has a narrow-linewidth well below 300 kilohertz. That is so perfect for coherent light applications.

As I mentioned, we have sampled to some key module manufacturers and getting great feedback. So we're in a qualification process for that, building upon the acquired product from HieFo. As for CPO, that's clearly is going to be high-density, low-power, high-bandwidth transport solutions, right, for future scale out and scale up. As I mentioned in the past, just the current way of the CPO, our FiberEdge product is going to be -- is not going to be -- it's not an opportunity for FiberEdge product because the customers will be using integrated solutions.

So we have to be developing application to participate meaningfully in this huge emerging opportunities. This so happened that the HieFo acquisition, the lasers we buy about and also the Gain chips and semiconductor optical amplifiers can really serve the part of the light sources for CPO scale-up applications. So we're working internally, we're also working with the partners to provide high-power DWDM CPO laser source solutions. This is more for 2028 opportunities. So you see our pattern. We're not only investing in the current generation, not only investing in the growth driver for the immediate next year, but the year after that as well. So we got a very healthy pipeline of new products and broaden our portfolio, serving the high-bandwidth AI data center connectivity space.

Awesome. It's very clear. I wanted to ask Mitch, I mean, guiding for, I think, more than $10 million sequentially in OpEx. Presumably, that's not necessarily all going to snacks for the break room, but maybe you could just talk about you mentioned R&D spend and SG&A was coming down as a share of total sales. But where are your focus points? Is it going to hiring? Is it going to -- are there maybe non-CapEx type expenses, prepayments, anything like that? That would be helpful.

Sean, we expect to continue to invest in the business. We've said that, that's kind of our #1 use of capital -- in capital allocation. And the vast majority of the incremental spend is going towards high conviction R&D programs, largely in data center, but also supporting LoRa, and it's not in SG&A. So maybe just allow me to add some percentages to that color, right? So R&D in Q1 -- the Q1 that we just closed, R&D was 17.6% of net sales. That's 20 basis points higher than a year ago. and up 17% year-over-year. But when I compare that to SG&A, in this Q1 that we just closed, SG&A was 15.1% of sales. That's a decrease of 200 basis points, right?

So -- and SG&A as a percentage of sales has been in a steady decline, and we project this decline to continue in Q2. So our OpEx growth is really grounded in organic investment in our core products, data center and LoRa. In terms of the composition of spending, there's hiring and there's project spend as well. But really, I think the key is that we see some very good returns on this R&D investment.

If I may double-click on the 15% SG&A is actually sales and field application engineering are increasing sequentially. So what gets squeezed or getting more operational leverage is SG&A cost.

Our next question comes from the line of Christopher Rolland with Susquehanna.

I want to echo my congrats as well. I did want to double-click on optical, perhaps a question for you, Hong. If you could talk a little bit more about the contributors to growth there, in particular, how much of this is LPO? How much of this is a single module maker versus a broad 1.6T deployment? And any update on timing for CW and SiPh chips would be great as well.

Thank you, Chris. Thank you for your questions. For Q1, the predominant data center revenue is 800-gig FRO and LPO. We do see a clear trend of sequential increase from a mid-single digit reported in Q4 for LPO, we're seeing sequential increase and the 800-gig is the main driver for the revenue for Q1. Into Q2, we're going to be having the early -- the Q1, we also have the early ramp of the CopperEdge revenue to support the second half of significant ramp in the cable demand. So in Q2, we are going to be seeing the continued strength of 800-gig FiberEdge for both FRO and LPO. We're going to be continuing to see the ramp of 1.6T CopperEdge to support ACC for the second half. And we're going to be for the first time seeing the FiberEdge revenue to support 1.6T optical transceivers in the new DSP designs.

As for the revenue source, it's very broad-based. We, as I mentioned, have qualified with almost all module manufacturers, established and emerging ones, and we were able to serve some drop in demand even in Q1 because they were not in the forecast, and they are very important strategically for us to go forward, and we wanted to support their initial volume even within -- well within the lead time.

Yes, sorry, go ahead...

No, no, no. Hopefully, that answers your question for those.

Yes, that did. Maybe switching to the copper side, I think. You answered a bunch on ACCs. I guess my only question there is maybe diversification beyond your main customer. How is that progressing? And then in your presentation, you mentioned linear equalizer, I think, on PCB about active backplane applications. If you can speak a little bit more there and the opportunity, that would be great.

Yes. Thank you. So on the ACC diversification, we continue to having samples being evaluated by multiple hyperscalers and enterprise customers, and that reception has been going well. And as Rick early on asked, the MSA for ACC is going to help as well. And we're actually also seeing the exciting opportunities from multiple potential customers on the linear equalizer onboard. Those are for the real programs, and we have the hyperscaler engagement. We have their manufacturing partner, ODMs close engagement as well. So this is really going well. We do expect, as I said, from linear equalizer onboard on the backplane and also ACC cables, additional design wins in the coming quarters.

Early on, people have some doubt or speculation on copper is to the end of the life. We need to put a terminal value on it, just not yet. It is going not only strong, it's very strong on higher data rate as well beyond 224 gig. We're engaging with customers to support higher data rate as well. So we're very optimistic about CopperEdge product for its growth potential in the future.

Our next question comes from the line of Quinn Bolton with Needham & Company.

I also offer my congratulations on the nice results and outlook. I just wanted to follow up on Chris' question there on the onboard linear equalizer opportunity. You mentioned hyperscalers. But I was going to ask kind of from a signaling perspective, are you guys seeing demand for linear equalizers for both unidirectional SerDes on those backplanes as well as bidirectional SerDes. Can you just talk about your ability to support both UniDie and bidirectional SerDes on those backplanes? And then I've got a follow-up.

Yes. Thank you, Quinn, for that question. Yes, we did get request for BiDi linear equalizer as well. So our current CopperEdge portfolio can only support unidirectional and -- but our engineers have been put on the drawing board, we have a preliminary architecture formulated to be able to do BiDi on one chip and to support the future more compact interconnect topology.

And any idea when that might be ready? Would that be ready calendar '27?

So once the process -- once we define the product definition will go through the simulation stage. Clearly, BiDi integrated in one package need to be very carefully balanced so that they don't have the crosstalk and introducing additional noise, but it's totally doable. And we're just looking at the different competing priorities and getting that defined. So we are in the stage of engaging with the customers and trying to sync up with the model. And therefore, after that, we will allocate resources and start doing the development. The good news is the key building blocks of IP, we already have that in existence. The CopperEdge product, we have 3 generations and going for different noise suppression and different frequency domain equalization schemes. And we have rich building block IPs available to drag and drop and getting a product put together pretty quickly.

Got it. And then I guess, either for Mark or Hong, just you talked about expecting data center growth to accelerate if I've got my numbers right. It looks like with the 35% sequential growth in the July quarter, your data center business in the first half of '27 will be up about 62% versus the first half of '26. Would you care to give us some sense, do you think the entire data center business, could that grow 70%, maybe 75% year-on-year in fiscal '27? I think last quarter, you had expressed confidence in 50% year-on-year growth for data center.

Yes, Quinn, so you're right. In last quarter, that floor number certainly created some confusion. I will not let our growth potential or aspiration. I will not cap that for the year-over-year. As I said, Q2 to Q1, we are very comfortable about sequential 35% quarter-over-quarter growth. And based on the visibility, the backlog and the booking we have, we can comfortably say in that second half, the growth rate is going to be accelerating because of the additional growth drivers in 1.6T CopperEdge and FiberEdge, 1.6T and HieFo optical components. So it's going to be an exciting year in FY '27, and we'll be having the unprecedented year-over-year growth.

Our next question comes from the line of Cody Acree with The Benchmark Company.

Congrats on a very strong quarter outlook. Maybe, Hong, can you maybe just talk about your LPO and 1.6T optical expectations, whether that be through module partners? Or is that more hyperscale driven?

Yes. So Cody, thank you. The 1.6T, the current design wins we have from multiple major module suppliers and serving different end markets. They're either hyperscalers or a GPU company providing the system solutions. So the initial ramp is going to be FRO fully retimed optics. And we do have the linearized drivers and TIAs provided to the module customers. They're in an evaluation to do LRO or LPO at 1.6T. But the initial volume ramp is going to be FRO, and we do have the right product in the pipeline to be evaluated by our module customers.

And Hong, can you maybe help us frame the total available addressable market for both ACCs and LPOs as you exit this year and look into next?

Yes. So Cody, I think we will probably -- when we think that portfolio cleaned up, we're going to be having a Tech Day or Analyst Day. And at that time, we'll be providing more comprehensive TAM projection and -- for our broad product portfolio. So at this point, we are just aggressively attacking the opportunities ahead of us. We will be having more quantitative information provided in the future.

Our next question comes from the line of Tristan Gerra with Baird.

Given the commentary about Broadcom CPO ramping in volume not until 2029 and the potential for ACC to be at 800-gig per lane, how should we look at the duration of the double-digit growth that you see in ACC? Obviously, the business is very strong now. But do you think that, that business continues to grow 3, 4 years from now and how well positioned you are from a technology standpoint, particularly as we get new waves of switches that have stronger SerDes?

Yes. Tristan, thank you for the question. So yes, Broadcom, clearly, they're the industry leader. They provide outstanding SerDes quality. And with a good signal integrity, that has been the essential performance enabler for the linear redriver. And because if the signal come out not very good, linear redriver cannot make it whole a lot better. So we are the pure beneficiary of the Broadcom good SerDes rather than a victim. Their 100-gig, 200-gig and go into 800-gig per lane in the future, they can go for a certain distance, the so-called bump-to-bump link budget. The redriver -- linear equalizer can only make it better to stretch the reach to improve the signal integrity compared to the cables or trace without it.

So this is really a complementary capability. It's not like their better SerDes would not need a linear equalizer in a way, as you know, that the current ACC cable customer, they are using Broadcom SerDes as well. And we have our linear equalizer in ACC cable to support the interconnects. In their CPO is primarily my understanding, the current road map to support CPO scale-out. They are the leading switch provider anyway without a CPO or with CPO. So motivation for them to do CPO could be a little bit different from the other leader in the industry. That's for 800-gig per lane, and that is 2 generations ahead. And in general, as the data rate goes up and the transmission distance is shortened.

And the rack on the other hand, is going to be going bigger. So the topology will have to require a longer length. So you can do CPO to have the optical scale up, but you always have some traces. It's just a tool or a cue to use optical scale up. And by putting a redriver in between, you may as well just extend the reach well long enough to allow the copper scale up. So this is a complementary capability. I think it's going to be there for a long while.

Okay. That's very useful. And then could you talk about your medium-term views on LoRa? Clearly, it's rebounding. We're past the inventory accumulation that we saw 1.5 years ago. Where could this business grow in the next few years? How does that improve the mix? And also, any update on whether the deployments that you see this year and next year are primarily driven on the base station side or the actual sensor side? And I don't know if there's any numbers to back up versus base station product for LoRa?

Yes. So yes, there are several questions. Let me try to address a couple of them. First of all, the growth potential we see is really exciting. As we mentioned, the 3 things we did and increasing the bandwidth and increasing the LoRa capability by adding other RF protocols to unlock new application verticals like security, like a smart building and the mass market like Amazon Sidewalk. Those are all bringing new growth drivers to LoRa. The second question is really a good one. Tristan, you know this market well. And we see if a year ago, LoRa growth has been primarily on the end nodes. So we are increasing the end node right now to 150 million end nodes. And -- but started from last 9 months, 12 months ago, and our integrators has been commenting, they are adding new gateways. That means they have more sensor devices and they need to improve the coverage and capacity.

So that with the gateway increase, and they will be just pulling in more capacity for end use -- end node applications. So that's why we see this LoRa's growth driven by the 3 pillars is going to be sustainable. One wildcard is the Edge AI application. So it's an exciting opportunity, and we're just starting. So we have been helping the market and generating a lot of collaterals and making the market enablement effort to accelerate the growth on that. Thank you, Tristan.

Our next question comes from the line of Kyle Smith with Stifel.

This is Kyle Smith on for Tore Svanberg at Stifel. I'll also echo my congratulations on the really strong print and guide. I think in your prepared remarks, you mentioned that Gain chip demand currently exceeds your supply. So a bit of a 2-parter. One, it would be really helpful if you could kind of quantify the degree to which that demand exceeded the supply. And then secondly, I know you mentioned that your capacity expansion plan is well on track. But maybe when do you kind of expect your increase in capacity to intersect with the demand that you're seeing from the market?

Thank you, Kyle. The Gain chip, as you know, HieFo and its predecessor, EMCORE has been a leading supplier of Gain chip integrated into the tunable laser for metro and the coherent applications, we have a solid customer base. And after the acquisition of HieFo by Semtech, we have a lot of inbound interest from the customers that HieFo has not been able to serve. So you aggregate those -- the demand will probably outpace the capacity by about 3x or so. So we clearly are adding capacity by adding more shift by adding more manufacturing space, which is primarily clean room and by adding more process equipment. So we are doing very creative ways to expand the capacity. We expect by the end of this year, we can get the capacity increase by about 3, 4x and end of next year and it's going to be another 3, 4x so to serve the high-end coherent market.

That's really helpful color. And I guess pivoting over to LPO and LRO, I mean, it's pretty clear that the market is growing and the opportunity just continues to expand. So I guess from your position today, based on the outlook kind of that you provided on the call, to what degree do you feel that, that raised guide is coming from potential market share gains? And then to what degree do you feel it's more so coming from just the market as a whole expanding meaningfully as we kind of see these deployments take shape?

Yes, Kyle, I think the LPO, LRO, we see the growth is primarily driven by the customer shift from FRO to the LPO, LRO. The reason for that is that they evaluate their connectivity topology and to see where they can save power. And we do expect in a year or 2 timeframe, the linearized solution will account for 25% plus/minus of the total transceiver mix.

And our final question comes from the line of Craig Ellis with B. Riley Securities.

Congratulations on the strong results. I wanted to start off just with a clarification. So clearly, we're seeing much higher growth here than we were expecting 3 months ago with the business seemingly tracking in the 70s range now versus LoRa of 50% earlier. The question is this, where across TIAs, LPOs and ACC, are we seeing the greatest acceleration versus what we were expecting 3 months ago?

Yes, Craig, that's a great question. We were on the road right after OFC, the dynamic has picked up pretty dramatically. The demand across the board, not on just one product across the board for FiberEdge, CopperEdge, 800-gig, 1.6T LPO, LRO and lasers, they all have demonstrated such a strong demand. So it's broad-based.

That's very helpful, Hong. And then one of the things that's clear from your prepared remarks and here in the Q&A is just the significant visibility you have in the back half of the year. Can you characterize across the different product groups? To what extent does that extend? Are you actually getting visibility into fiscal '28? Or is it really at different points into the back half of this fiscal year?

Yes. Back half of this year into the first half of fiscal '28.

And we have reached the end of the question-and-answer session. Therefore, I'll now turn the call back over to Mitch Haws for closing remarks.

That concludes today's call. Thanks to all of you for joining us today, and we look forward to seeing you at various investor events over the coming weeks.

Good day, and thank you for standing by. Welcome to Semtech Corporation's Fourth Quarter and Fiscal Year 2026 Earnings Conference Call. [Operator Instructions] Please be advised that today's conference call is being recorded.

I would now like to hand the conference over to Mitch Haws, Senior Vice President of Investor Relations for Semtech. Please go ahead.

Thank you, and welcome to Semtech's Fourth Quarter and Fiscal Year 2026 Financial Results Conference Call. Participants on today's conference call are Hang Ho, our President and Chief Executive Officer; and Mark Lin, our Executive Vice President and Chief Financial Officer. Before we begin, I would like to highlight upcoming investor events, including the optical fiber communications conference sorting tomorrow and the Roth Technology Conference on March 23.

Today after market close, we released our other results for the fourth quarter and fiscal year 2026, which are posted along with our earnings call presentation to our Investor Relations website at investors.semtech.com. Today's call will include various remarks about future expectations, plans and prospects, which comprise forward-looking statements. Please refer to today's press release and see Slide 2 of the earnings presentation, as well as the Risk Factors section of our most recent annual report on Form 10-K.

For a number of risk factors that could cause our actual results and events to differ materially from those anticipated or projected on today's call. You should consider these risk factors in conjunction with our forward-looking statements. We will refer primarily to non-GAAP financial measures during today's call. We'll also be referring to results for our fourth quarter of fiscal year 2026, unless otherwise noted. Please see today's press release and Slide 3 of the earnings presentation for information regarding notes on our non-GAAP financial presentation.

The press release and earnings presentation also include reconciliations of our GAAP and non-GAAP financial measures. With that, I will turn the call over to Hong.

Thank you, Mitch. Good afternoon to all of you joining today. Semtech closed fiscal year 2026 with a significant momentum, achieving a record $1.05 billion in net sales, a milestone that reflects the progress we have made and the trajectory will be -- lies ahead. We drove strong sequential and year-over-year revenue and earnings growth, advanced our data center road map to capture compelling design wins opportunities and continue to optimize our product portfolio, all while executing on the R&D and expense initiatives, we believe positions Semtech for an exciting next chapter.

Looking at the Q4, net sales were $274.4 million, up 3% sequentially and up 9% year-over-year. For the year, revenues were $1.05 billion, representing annual growth of 15%, driven by continued strength in our data center and the lower portfolio. Adjusted diluted earnings per share were $0.44, up 10% year-over-year.

For fiscal year 2026, adjusted diluted earnings per share were $1.71, growth of 94% over the prior year. In addition to delivering strong revenue and earnings growth, portfolio optimization remains a key focus of execution. As announced earlier this month, we acquired HFO Corporation, which represents an important strategic building block for Semtech. Tipo is a California-based manufacturer of high-efficiency indium phosphide-based optoelectronic devices, including game chips and the CW laser chips that are critical components in the optical transceivers are in today's data centers. Put simply, [ Hifumix-lighting ] meeting building blocks that site a hard on high-speed optical entertainment.

The strategic rationale is straightforward. -- as data center architectures evolve to 1.60 and 3.2T. -- the complexity of optical interconnect increases dramatically by bringing HieFo's proven indium phosphide laser technology, together with the Semtech's industry-leading TIAs and laser drivers, we can codevelop and co-optimize performance across the laser modulator in driver interface. And increased some tax content opportunity from high single-digit dollars in an 800 gig module to about $80 in a 3.20 module.

We believe this combination will result in a more integrated, more efficient ship set, one that reduces system power consumption and gave the hyperscaler a differentiated solution for high-bandwidth optical transceiver modules. We have developed a comprehensive investment plan for our Hamburg, California facility, to expand domestic capacity and accelerate the product development, integration of HieFo into Semtech's operation is underway, with the transaction expected to be accretive to non-GAAP diluted earnings per share within the first year.

We are generally excited about the people and technology joining Semtech and we see significant untapped potential for high efficiency lasers in different interconnect applications.

Finally, we continue to make progress on the divestiture of the cellular module business, and we are increasingly encouraged by the level of interest and engagement. We remain confident this business represented a compelling opportunity for the right acquirer, and we remain focused on bringing this process to a successful conclusion.

Now let me move to a discussion of our end markets. For Q4, infrastructure net sales were $86.3 million up 11% sequentially and up 25% year-over-year, strongly supported by our data center business. For fiscal 2026, infrastructure net sales were $310 million, growth of 27% over the prior year. For Q4, net sales of data center were a record $63 million, up 12% sequentially and up 26% year-over-year, benefiting from strong demand for our broad portfolio, including our market-leading fiber edge -- is whose net sales set another record.

In Q4, we also started shipping into LPO transceivers with revenues in line with the outlook we provided on the last quarter's earnings call. year 2026, our data center revenues were a record $223 million representing an annual growth of 58%. Our optical and copper product lines are firmly established with hyperscalers as a differentiated and high-performance offering.

Power efficiency has become one of the defining constraints of the modern AI infrastructure. As hyperscalers major data center capacity in megawatts the ability to move data faster while consuming less power at the networking layer is no longer just a differentiator ESMenabler.

Our analog solutions address this directly enabling operators to scale next-generation architectures at 800 gig, 1.6 and eventually 3.2T. Demand for 800-gig PIA solutions remains strong and broad-based with an increasing momentum throughout 2026. At 1.6T, we are engaged across a wide range of transceiver programs and expect volume ramps to build as hyperscalers roll out their new XPU and switch platforms using reduced power of 1.60 transceivers throughout the year.

On LPO, design wins with several leading U.S. hyperscalers validated our TIA and driver solutions in 800 gig transceivers. We are very excited to be a member of the new XPO MSA to define specific patients and enable high bandwidth, high density and low power switches. By combining with the low loss copper entertainment, such as flight over wires, our linear redriven PCB traces, some hyperscalers are becoming increasingly bullish on 1.6T LPO instead of using LRO for the first layer of the scale-up fabrics.

We continue to expand our LPO IC portfolio with 1.6 LPO drivers and TIA is expected to come to market this year. In supporting further proliferation of low-power linear optics, Semtech, along with other industry leaders are developing MPO, our near package optics MSA for low-power, high-density and high-bandwidth solutions.

Successful deployment of 800-gig LPO transceivers gives hyperscaler confidence in MPO at the next evolution of the optical solutions. We are excited by the increased content available to Semtech in MPO deployments. Active copper cable, our ACC continues to gain significant traction. Customers evaluating ACCs against the incumbent solutions are seeing compelling performance advantage in the form of a robust linked margin and transformative power savings versus DSP-based solutions.

Alongside our cable solutions, customers are increasingly evaluating our corporate leader equalizers for onboard integration to enhance signal integrity across high-speed links and opportunity, we are confident we'll convert into design wins over the coming quarters. One of these use cases is active back point using copper edge ICs, which our cable partners will demonstrate at OFC.

Finally, we co-authored the ACC MSA, helping establish ACC technology as a leading solution for low-power and high-performance copper links. Members of the MSA span IC, XPU and cable suppliers, or in partnership with major hyperscalers. We believe the ACC MSA accelerates adopts and curve for the entire industry. By establishing common specifications, we reduce fragmentation lower deployment risk for hyperscalers and make it easier for the ecosystem to develop around ACC as a standard, not just the proprietary solution, that is good for customers, is good for Semtech.

We look forward to seeing many of you at OFC starting tomorrow. This year, we are showing live demos across several of our key product areas. They tell a clear story about where Semtech is positioned in the data center interconnect market. Starting with copper. We are demonstrating 1.60 ACCs running live traffic to adidas 224 gig [ SerDes. ] We also show in next gen recent 448 gig per channel copper edge chips as the AI clusters scale -- the demand for low power and low latency copper interconnect continues to grow, and we think we are very well positioned to meet the market.

On the optical side, will be demonstrating NVIDIA's 1.60 DRA transceivers powered by some tech TIAs and laser drivers, running NVIDIA Switch platform, will also be demonstrating 100 Ethernet, which running live traffic over both single mode and multimode fibers supporting FRO, LRO and LPO configurations across the [indiscernible] 6 platform, our field on Semtech silicon.

We are thrilled to demonstrate the breadth of our optical portfolio across a multi-vendor ecosystem. We will also be demonstrating our next-gen recent 448 gig per channel modulator drivers and TIAs addressing increasing bandwidth demand to support future generation AI workloads. We're also showcasing our indium phosphide CW laser and game chip technology for tunable laser applications with outstanding power efficiency over temperature performance and the far field beam profile products from our type of acquisition.

Across copper and optical and both near term and next generation gives us a tremendous opportunity to show Semtech's position across the full hyperscale interconnect step. We believe we are positioned for multiyear growth opportunities supported by our expanded portfolio. We expect to start shipping copper Edge for the 1.60 ACC hyperscaler deployment this quarter, with demand accelerating throughout the year. We also expect fiber edge design wins for 1.60 transceivers with a significant ramp in the second half of the year.

Additional revenue growth drivers in the near future are expected from additional design wins for ACC and other hyperscalers, linear equalizers onboard across multiple customers. gate shifts and CW lasers in transceivers and our market-leading 400-gig FiberEdge and copper edge products for 3.2 interconnect solutions.

Given the breadth of our data center portfolio and designing traction across the expanding set of customers, we expect data center year-over-year revenue growth this fiscal year to exceed 50%.

Now moving to our high-end consumer end market. Net sales for Q4 were $36.6 million, down 13% sequentially and up 3% year-over-year. Net sales for fiscal year 2026 were $155.1 million, up 5% year-over-year, driven by both our TVS and per se product portfolios. Consumer TVS revenue continues to ramp well ahead of handset volume growth, and we expect another year of revenue and design win momentum.

We expect consumer TVS revenues to increase next quarter, a function of improved seasonality and share gains at a leading handset manufacturers. In addition, per se continues to broaden its design win footprint with adoption expanding across smart glasses and smartphone platforms in supporting both current and upcoming product launches.

The integration of the force sensing portfolio is progressing well with the initial product shipments already underway. Customer engagements and design win activities continue to build and we are increasingly optimistic about the cross-selling opportunities. This combination unlocks across our combined customer base.

Moving to our industrial end market. Q4 industrial net sales were $151 million, up 3% sequentially and year-over-year. With another solid quarter for LoRa, for the full year, industrial revenue was $584 million, 13% over the prior year. LoRa enabled net sales were $39.6 million, in line with the Q3 and up 7% year-over-year, supported by continued expansion across several application verticals such as smart utilities, smart beauty, smart city and asset management.

For the full year, LoRa revenues were $156 million, representing full year sales of 34%. We had a strong presence at the CES this year, showcasing new LoRa application use cases, 4 themes highlighted in our presence, AI integration multi-protocol connectivity, global network expansion and the convergence of the industrial and consumer IoT together, reflecting a technology that is broadening its reach across a growing range of markets. HII emerged as another defining trend as sensors increasingly process data locally for latency, privacy and bandwidth reasons, our collaboration with an ecosystem partner demonstrated how LoRa on and I work together to enable predictive maintenance in industrial environment.

Demand for a solution that combines LoRa with multiple protocol flexibility is accelerating. In order to facilitate LoRa adoption, we recently signed an agreement with a technology partner to support software development to activate and support other protocols. We are rolling out a Z-Wave first with a Z and a threat and matter to follow. We expect beta units will be available to deployment partners in Q2 of this year, as multiple protocol smart home and security solutions.

The markets moved towards a cynical skill solutions is exactly what LoRa is be starting to address reducing complexity for customers while expanding our addressable markets. Customers who purchase LoRa plus transceivers now get royalty free access to an SDK and development tools, silicon and software together from a single source.

Additionally, Amazon and Ring announced a new line of LoRa powered sensors spending security, safety and home control applications, all operating on Amazon Sidewalk in plans to launch this product in the U.S. in March, followed by expansion across Canada, Mexico, Europe, Australia and Japan. This demonstrates lower readiness from mass market consumer adoption at Amazon scale, a significant evolution from this industrial and commercial routes.

The ecosystem continues to scale, now spending over 125 million LoRa and connected devices across 70 countries well beyond the early adoption and into mainstream deployment. With lower technology, we now have established 3 pillars of low-power connectivity platforms. lower on, Lowes with multiple protocols and Amazon sitework.

With this, those vectors, we believe LoRa long-term drill rate to be approximately 20% on and quarterly sales to reach from $35 million to $45 million. Our IoT systems and connectivity business recorded Q4 net sales of $89.9 million, up 2% sequentially and down 3% year-over-year.

For fiscal 2026, we revenues were $354 million, up 9% compared to last year. We continue to bring products to market that address gaps in how industrial customers connect remote infrastructure. In Q4, we launched the AirLink RX400 and the EX400 the industry's first rugged 5G RedCap routers, purposely built for mission-critical industrial deployments. These routers delivered 5G performance with less than 1 watt of the idle power roughly 1/10 of the draw of standard 5G equipment, making solar and battery backed deployment practical for the first time. This allows our utility oil and gas and transportation customers to operate in remote locations where great power is not always available. Customer engagement at the distribute pack in February reinforced our conviction that this product is well timed to address a real market need.

Looking back on fiscal 2026. I'm proud of what the team has accomplished. We delivered strong revenue and earnings growth through disciplined executing a differentiated portfolio and a relentless focus on the customers and the markets where Semtech can win. This was not just a strong year financially. It was a year in which we fundamentally strengthened Semtech foundation.

Looking at where we stand today, I'm more confident than ever in our positioning. The AI data center build-out is one of the most significant infrastructure investment in a generation -- and we believe Semtech is well positioned with a broad purpose build portfolio of solutions designed for 1.6T and 3.2T error. We believe our continued investment in our core assets through R&D and acquisitions helps ensure we are not just keeping pace with the next-generation technology. we are helping to define it.

And importantly, we now have the financial flexibility to diligently evaluate and pursue the strategic investments that will accelerate our pets. We entered fiscal 2027 with the momentum with a clarity of purpose and with the stronger Semtech than we have had in years.

I want to thank our employees, our customers and our shareholders for their continued confidence in us with just getting started and the opportunities ahead has never been more compelling. Our key focuses for 2027 fiscal year include: One, accelerating business growth by supporting customer reps with sufficient availability and strong operational metrics as we compete in a capacity-constrained environment; two, intensify R&D investment to add new drug drivers and solution differentiation by maintaining diligent governance of R&D investment with the goal of driving customer wins and delivering strong financial returns; three, transform tech by strengthening our winning culture and making major progress in portfolio optimization.

With that, I will now turn the call to Mark for additional details on our financial results and our outlook for the first quarter of fiscal; 27. Mark?

Thank you, Hong. For Q4, we recorded our eighth consecutive quarter of net sales growth with record net sales of $274.4 million, above the midpoint of our outlook and up 9% year-over-year. For the fiscal year, net sales were $1.05 billion, up 15% year-over-year. Net sales trends by end market, reportable segment and geographic region are included in the accompanying earnings presentation.

Adjusted gross margin was 51.6%, above the midpoint of our outlook. Total semiconductor products gross margin was 61.7%, up 40 basis points sequentially and up 350 basis points year-over-year. Total semiconductor products gross margin was above the high end of our outlook range. The result of favorable mix from our LoRa and data center portfolio. We expect gross margin contributions from new data center products from our copper and optical 1.6T portfolio ramping in the second half of this year will be accretive to both our semiconductor products and Signal Integrity Products gross margin.

IoT systems and connectivity gross margin was reflective of mix related to net sales growth and cellular modules with Q4 at 31.6%. Adjusted net operating expenses were $91.5 million, slightly above the midpoint of our guidance range. Adjusted operating income was $50.0 million. Adjusted operating margin was 18.2%, adjusted EBITDA was $57.4 million, and adjusted EBITDA margin was 20.9%, with all of these metrics above the midpoint of our guidance range.

Reflective of capital structure changes, Semtech was in a net interest income position in Q4 at $0.1 million, which reflects a sizable change from the $11.2 million of net interest expense reported a year ago. For fiscal year 2026, adjusted net interest expense was $11.5 million compared to $70.6 million in fiscal year 2025.

We recorded adjusted diluted earnings per share of $0.44, above the midpoint of our guidance and full year adjusted diluted earnings per share was $1.71. Operating cash flow for Q4 was $61.5 million, sequentially up 30% from $47.5 million and up 84% and from $33.5 million a year ago. Free cash flow for Q4 was $59.1 million, sequentially up 32% from $44.6 million and up 91% from $30.9 million a year ago.

Operating cash flow and free cash flow for the stand-alone fourth quarter each exceeded the amounts reported for all of fiscal year 2025, driven by improvements in both our capital structure and operating performance.

Strong cash flow generation has allowed us the operational flexibility to invest in R&D projects as well as to invest strategically via tuck-ins. The increased R&D investment in our core data center LoRa and sensing portfolio has yielded strong returns. The aggregate consideration for the sensing portfolio we acquired in October 2025 and the laser and gain ship business we acquired in March 2026 is less in the free cash flow we generated from Q4.

I believe our balancing of R&D spending, along with prudent use of capital for capacity expansion and acquisitions positions us to generate meaningful long-term returns for our shareholders. We ended Q4 with a cash and cash equivalents balance of $195.2 million and debt was $503 million, unchanged from last quarter.

Our adjusted net leverage ratio was 1.3 as of the close of Q4, down sequentially from 1.5 and down year-over-year from $2.3 million.

Now turning to our outlook for the first quarter of fiscal year 2027. We currently expect net sales of $283 million, plus or minus $5 million, up 13% year-over-year at the midpoint. We expect net sales from the infrastructure end market to increase sequentially, supported by projected sequential data center growth of 12%, including initial copper age production shipments supporting a hyperscaler at the tail end of the quarter.

We expect net sales from our high-end consumer end market to increase about 9% sequentially or about 13% year-over-year, benefiting from improved seasonal trends, market share gain in our TVS products and contributions from the Force Sensing portfolio we acquired in the fourth quarter. We expect net sales from our industrial end market to be about flat with LoRa increases offsetting decreases in IoT systems and connectivity.

Based on expected product mix and net sales levels, we expect adjusted gross margin to be 52.8%, plus or minus 50 basis points. Our gross margin outlook for our total semiconductor products is expected to be 60.4%, plus or minus 50 basis points, down 130 basis points sequentially and include initial ramp costs from the HieFo acquisition.

Adjusted net operating expenses are expected to be $96.9 million, plus or minus $1 million, resulting in adjusted operating margin at the midpoint of 18.6%. Included in the higher first quarter adjusted operating expense outlook, our R&D costs associated with the addition of the Force Sensing business, along with increased investment support of our growing data center portfolio, including the HieFo acquisition.

We have demonstrated strong returns on our R&D investment and expect incremental returns from these investments. Adjusted EBITDA is expected to be $59.5 million, plus or minus $3 million, resulting in adjusted EBITDA margin at the midpoint of 21%. We expect adjusted interest and other expense net to be approximately $0.5 million.

We expect an adjusted normalized income tax rate of 72% for all of fiscal year 2027, an increase from 15% in fiscal year 2026 due to a geographical shift in pretax profits. These amounts are expected to result in adjusted diluted earnings per share of $0.45, plus or minus $0.03 based on a weighted average share count of 96.6 million shares.

Thank you, Mark. We can now turn the call back to the operator for the question-and-answer session.

[Operator Instructions]. Our first question is from Sean O'Loughlin with TD Cowen.

Congrats on the solid results and continuing to execute on your strategy. I wanted to ask a question on the HieFo acquisition. I think the strategic rationale you laid out is pretty straightforward. But wondering if you could expand on the initial applications you're targeting, it sounds like mostly transceiver side.

And then maybe more importantly, where are you for the Indium phosphide-based products in the qualification or design cycles at some of your customers? And what should be our expectations for when some of that starts to fold into the to model in a material way?

Yes. Thank you, Sean, for the question. So the HieFo product right now in production is a game chip use in tunable lasers. Those are the high-end lasers to support the transmission over 80 kilometers, 40 kilometers for metro and data center interconnect applications by leveraging about 4 decades of laser design experience and they have been the leading supplier for gaining chips for IPL.

This is a product already in volume production and demand is increasing. And right now, it's capacity limited. So this product for 2027 probably will be contributing roughly about a high teens level of revenue. But we are adding capacity as we speak in the second week and a proud owner of this asset.

We have also demonstrated will showcase tomorrow at OFC, the intensity modulated direct drive lasers our CW lasers used in optical transceivers. This laser is really well designed to support high conversion efficiency. For example, well plug efficiency at the room temperature the 42%, that is much higher than a typical about 25%.

Then over temperature performance is outstanding and the [indiscernible] profile is outstanding. So that make it much easier to couple into single modfiberour silicon photonics waveguide. And we do not have enough capacity right now to support the industry demand after the announcement, you can just imagine, we've got multiple customers reaching in and asking for samples. So we will be able to provide samples to the key customers to support the evaluation in when we are building capacity to support the future is.

As for putting in the model, probably once we have a better understanding on the equipment lead time and the capacity in the future periods, we will provide more guidance on the revenue contribution from laser CW lasers. Right now, we're planning to intercept at 3.2 transceivers, of course, it can support 1.6T transceiver as well.

Great. And I appreciate all the color there. If I could just ask a real quick follow-up maybe for Mark on that topic. How should we be thinking about the CapEx line as you talk about those capacity expansions at Alhambra, I guess maybe Hong alluded to the answer to this question in his previous comment about understanding equipment lead times, but how everybody in the indium phosphide industry seems to be ramping capacity? And are you guys, I guess, for lack of a better term, at the back of the line there?

Yes. So the CapEx intensity is moderate. And the fab already have the key capabilities. Right now, we just need to selectively adding capacity in some areas. So the -- I know the whole industry is ramping up in equipment and equipment fab equipment in high demand as well. but we will be very creative in combined with a new and used market and to look at the fab and test equipment. The intensity, I would say, of the CapEx can be easily supported with, say, for example, 1/4 of the free cash flow as we have demonstrated in the past quarter.

Yes, Sean, the ramp is over multiple quarters, but I think Semtech's world-class operations team is on it. We really had the planning during diligence and the other good news is all of this CapEx, I believe, is qualifies for 48 investment tax for at 35%, and then there's some additional government programs that are available, especially to strengthen U.S. semiconductor manufacturing capacity.

Our next question is from Tore Svanberg with Stifel.

Hong, Mark, congrats on the results. So my first question is on copper Edge. So as that business now starts to ramp in fiscal '27, can you give us any sense for how big that could be, maybe tens of millions in dollars. And if you can't give us the size perhaps you could at least give us the mix between actual ACC cables versus linear equalizers.

Yes. So Tore, thank you for the question. So we are supporting the ramp, we're getting forecast and we're getting the materials ready as we gave the guidance in our prepared remarks towards the end of this fiscal year -- well, this fiscal quarter, in April time frame, we'll start shipping to the cable manufacturers who are our direct customers to support volume ramp in the mid of the year. So everything is on schedule.

As for the mix between DC and AC is still -- they are in the process of right now doing the direct level testing system validation. So we will be getting a better idea probably by -- in a month on the support. But we are getting the long-range forecast so that we can get a wafer starts in the fab and prepare the material readiness, as I indicated, that's a very key focus for our operations. So at this point, it's still too early to tell. But as for Linear Ecolizer on board, there are multiple customers supporting our activities and the -- I do believe that some of those based on the level of engagement, it will be getting qualified within the coming quarters.

So just on the leader equalize, the ACC side, the MSA establishment with 112 founding members our tasks and at work already started drafting the specific cases. This is still the early stage of the ACC in some customers experiencing the different customers experiencing different performance. It is very important to have this MSA and to drive the specification and educate the industry that what to expect with a certain level of the reach and gates and data rate.

So with the common understanding of MSA I do believe this technology is going to be proliferated much faster. So we have already seen some other hyperscalers lining up to getting their reg designed by using ACC.

Yes, that's very helpful. And as my follow-up and maybe more of a clarification question. So you talked about lower growing 20% longer term. But then you talked about a $35 million to $40 million quarterly run rate. You just did $40 million. So what exactly is that 35-45 range? I mean, are you expecting some volatility this year? Or yes. If you could just clarify what you meant by that.

Yes. Thank you, Terry. So historically, we have been seeing our trend of LoRa revenue has been increasing. But quarter-to-quarter, it's a little bit bumpy due to the project-based deployment of demand. That's why we give a range of plus/minus $5 million. But you see that sliding scale continue to go up with the center point. So I feel like I have never been this excited about our LoRa strategy. It's really with the dual-band to increase the bandwidth to address edge AI applications with LoRa plus to really get multiple applications in one SKU with the software and hardware are supported by us, and with Amazon, the sidewalk in the mass market.

So adding the value-added market we at the mass market really, we got multiple cost drivers. So I think that's why giving us conviction that 20% growth rate is very doable. One we can help the the ecosystem to adopt our LoRa plus protocol faster, we expect the X3 to accelerate.

Our next question is from Christopher Rolland with Susquehanna

I guess my first one is on the Indian phosphide laser acquisition. I guess, first, a clarification. This is all going to be internally manufacturers materials. I just wanted to clarify that.

And secondly, if you could talk about maybe your go-to-market strategy here and maybe even some revenue synergies with some of your other parts, are you going to kind of bundle this with fiber Edge? What -- are you going to perhaps go direct to hyperscalers? How are you going to approach this market?

Yes. Thank you, Chris. That's very good questions. To answer your first question, yes, the fab we acquired is vertically integrated, namely, we do the epro of epi wafers, we process wafers. We test it internally, and we do use the ecosystem on back-end packaging to increase the capacity. So that part, we are doing really well. This lasers because of the move real process -- so we do internally, that's how we are able to get the superior performance in conversion efficiency and over-temperature performance.

As for your second question, go-to-market strategy, you are absolutely right. We know that 100 gig per lane in transition to 200 gig per line that already put so much stress into the ecosystem. So we're really challenging the device designers with the performance margin. And when we evolve the data rate to 400 gig per mine, they really not hold a lot of margin to give out. So the codevelopment and co-optimize is so key in order to get the best electronic component with the best after electronic component. So now we own 2 sites of the question. So we'll be able to make a match to provide the best integrated solution.

And when we have that, we will provide the chipset with the reference design to our customer base. This way will be helped to accelerate the time to market for them and in return, will make our components more sticky. Right? So you use this kind of the electronic component like a TIA, a laser driver, modulator driver using our optical components, you will pretty much get a guarantee to work to deliver 400 gig performance. That's why we're saying we anticipate the major cut-in point is 3.2T transceiver modules. It could be earlier than that.

That's great. That's a fantastic strategy. One question I get from investors is around the eventuality of CPO, particularly for scale up Jensen here GCC was talking a little bit about the coexistence of both copper and optical in the rack, but some pushback I get from investors is around the role of copper, the closer that optical moves to the ASIC.

Obviously, you're expanding your market with lasers here into the CPO world. But perhaps you could also address that kind of pushback on copper where you see copper playing a role, not only in the next 2 years, but all the way through 2030, for example, even as you move CPO to the ASICs.

Great. Thank you, Chris. I listened to Jensen's call as well. So basically, what I was trying to say over the last year that he clarified it so well. So copper scale up is always still to be there. And CPL scale up is only more -- making more sense in the multi-rack systems. For example, he was talking about Albon and the 576 across 8 regs that way, because you're using active copper cable, it's very hard to get regs our point-to-point interconnected.

So you use the first opportunity, the signal out of the XPU to convert into optical. Optical not only be able to interconnect within rec and -- but it can also interconnect between regs.

The same thing for the [ Cibra ] platform, I was talking about NB1 that's also an 8 reg system on the NVL144. And that the CPO scale-up makes sense. So in general, I think the copper scale up is going to be the mainstream. It's going to be primarily used for within the rec the CPO scale up is going to be used in multiple reps. So don't put a terminal value on the copper debt.

And the industry also are formulating an NPO near package optics as a complement to CPO. So this way, CPO is one company thing, right? but NPL can define a specification to have specific geometry the IO pinout and keep our zone to leverage the entire ecosystems innovation to make it more scalable, more affordable.

So when in a CPO, we may not have much content in there, except for lasers, but in NPL, we will be all over the place with the laser drivers and TIAs and lasers and even silicon photonics modulator, as I mentioned, that is a very natural expansion of our portfolio by internal development.

Our next question is from Tristan Gerra with Robert W. Baird.

You've talked about the rising interest from hyperscalers about -- could you talk about the recent RECAP announcement and what it does to the LPO ecosystem? And any way to quantify what the ramp is going to look like medium term and when you expect the big inflection point in LPO revenue.

Great. Thank you, Tristan. So XPO, Arista just released that MSA. We are a very active member of the XPO defined high-density, low-power provide high reliability MSA for front panel switch, for example, to keep the same relics, but they will be able to collapse the form factor from essentially 4 RU into 1 without giving out any capacity. So it basically removed the packaging overhead because of liquid cooling is available, they can develop the cold plate so that getting between the optical transceivers.

And that is just great. And really that build upon the confidence on LPO and built upon the innovation from the entire ecosystem with multiple module manufacturers. NPL on the other hand, is a little bit more involved because that's involved the development of common specification on, say, Geometry [indiscernible] the IOs and the short line configurations -- so essentially, MPO is in a way XPO onboard. XPO is high-density package on the front panel of the box. In both configurations, we're going to be having a lot of content in there.

Again, the laser and the modulator drivers, the modulator itself and I -- and so we welcome this type of MSA. It's really going into our direction.

Okay. That's great color. And then for my follow-up, you've mentioned in the past some ACC opportunities on board. Could you provide maybe a little bit more feedback on what that is, what exactly is the use case for that? And how meaningful that could get over time?

Yes. The copper edge line Ecolizer on board. Our customers are defining 5, 6 different use cases in all to utilize the redrive capability to extend the link and improve the linked budget. So it can be on the switch, can be on the merchant DPU Board can be on the ASIC Board of hyperscalers, can also be on the backplane in an active back point by our cable partners.

Our next question is from Quinn Bolton with Needham & Company.

This is Robert on for Quinn here. Congrats on the quarter. Just 1 on you've been doing -- you've been active in doing acquisitions and have expressed intent to increase R&D with some of your capital up into this point. But any updates on potential divestitures -- last we heard, I think, ongoing and roughly in the third or fourth inning. So any update on that process would be great.

Yes, Robert, thanks for calling in. I'm not really good at sports analogies, but I think that we're making good progress. I'm optimistic in a good conclusion. At this point, I'd say that maybe incremental from last quarter, the the interested parties are spending some dollars on external consultants, so financial due diligence and on legal costs.

So we're at that stage. And when when there's additional kind of skin in the game, I think that does point towards a successful conclusion of the cellular module divestiture in the near term.

And just as a follow-up, it sounds like there are many tailwinds coming across for LPO as well as copper -- and I think last week, we kind of heard them ramping kind of in a similar time frame this year. Can you just refresh that for us? Is it sounds like ACCs could be ramping a little earlier now -- and then which do you see as the larger revenue opportunity over kind of the coming 12 months?

Yes, Robert, so we gave the time line ACC ramp. It's on time, on schedule, and then the 1.6T fiber edge product. And then beyond that, as I said, could be leading the ACC with other hyperscalers, linear equalize on board and lasers and 400 gig. So we like them all. And at this point, it is too early to tell which one is the bigger opportunity develop. We like them all and are going to be contributing pretty significant ways.

Our next question is from Joe Moore with Morgan Stanley.

You talked about 1.6 starting to grow in your GI business and other places. Can you talk about the line of sight that you have to 800 still growing? I assume it's still growing for now and you're layering in the higher speed on top of it. How long will that persist? And what point does it start transitioning over more?

That's right. So Joe, we did mention in the prepared remarks, the 800 gig is foundation -- the growth is strong. Demand is strong and broad-based. So that's the given that goes very strong, at least throughout FY '27. And as for the industry rolling out new XP use and then out the IO goes to 200 gig. So they're going to be evolving into 1.60 and in a low power 1.60 optics we are actually very well positioned even better than 800 gig.

Our next question is from Craig Ellis with B. Riley Securities.

Guys, congratulations on both the execution and a really strategic looking acquisition. Hong, I think near the end of your prepared remarks, you commented that you thought data center year-on-year revenue this year could be up around 50% one, just confirming that I heard that right? And two, can you help us understand what the biggest growth drivers are or rank the growth drivers that you see driving that to great growth.

Yes. Thank you, Craig. Yes, you heard it right. Year-over-year, we expect the data center revenue to grow 50% this year. So the driver is with hyperscaler 1.6 FiberEdge second half of the year and then there might be even leaner equalizer onboard contributing to the growth.

That's helpful. And then, Mark, congratulations on the strength in YouTube Hong on the strength in semis gross margin and what we're seeing in signal integrity, it's really nice to see those mid-high 60s levels reobtained. The comments on accretive second half products, are we signaling that gross margin for the segment could start with a 7 handle later this year or next year? Or would you still expect to be in the 60s?

So the -- the new products that we introduced do have accretive gross margins. So I'm very pleased with that. It's a great return on the R&D that we put into the products and they do contribute to growing gross margin. But -- we still also have other products, so at 800 gig, right? So I'm not expecting it to be in the 7s, but firmly in the ZIP code.

Yes. Well, congratulations on where it is.

Our next question is from Cody Acree with Benchmark stone.

Maybe just following up on Craig's question. You didn't mention LPOs among the drivers of that 50% data center growth. Any reason for leading that out? And maybe can you just talk about the breadth of expansion of your engagements with ACC and LPOs over the last few months and how you expect that to those to progress over the course of the year.

Thank you, Cody. So yes, the LPO continue to grow and as we said that Q4, we delivered just as we gave the guidance, mid-single-digit and signal the first ramp and to continue to ramp on the LPO, but we got them in the fiber edge product category, so included in there. The LPO adoption is proliferating, and that's also built the confidence for this XPO and NPL strategy. Then ACC is also getting more and more accepted better industry.

I mentioned that the hyperscaler -- 1 hyperscaler started ramping mid-year. And then more hyperscalers are embracing this ACC as well. Linear equalizer with multiple customers, the engagement has been going really strong. So I do believe that we not only have drivers lined up for this year and have multiple drills drivers lined up from future periods beyond FY '27.

Great. Mark, any thoughts on your OpEx trends for the year?

Yes. So I'm pleased that we're able to have the flexibility to invest in the business. Our outlook for next quarter in OpEx growth is really R&D. So I believe we've demonstrated strong returns on our investment. And our investment in R&D certainly yields stronger returns than interest expense, for example. But we do remain disciplined in our R&D investments, focusing on our core data center portfolio on LoRa and per se. And the 2 recent acquisitions, do you have some incremental R&D, but these are in our core portfolios.

Our last question is from Scott Searle with ROTH Capital Partners.

Just real quick on LoRa. I wondering if you could provide a little bit more color in the past. China was such a big percentage of the mix. I'm wondering how it was in the quarter and the pipeline of opportunities there. We continue to diversify away from the Chinese marketplace. And as part of that, Sidewalk has cropped up from time to time as being a large opportunity. It's been, I think, slow to date. I'm wondering if you could calibrate when we might start to see that contributing in a more meaningful fashion? And as a follow-up, -- just on the protection and sensing side, I think you indicated that there was a large win that kicks into the first quarter. I wondered if you could just clarify that and provide any additional color.

Okay. Thank you. Laura, right now, the gas is broad-based across multiple regions. China certainly is a good driver. And in Europe and in North America, as is equally strong. install benefited from multiple growth drivers out there. So sidewalk, certainly, it had a little bit of fall start several years ago. And -- but at OFC, not OFC, in January, they had over 10 product demos, all embedded with LoRa chips inside.

So they're going to start to deploy in March in North America. -- had a clear plan to proliferate into other countries. So we will see this time, but that can be a pretty significant opportunity for Semtech, so I will let Mark address the TVS question.

Yes. So TVS, we're seeing growth above, let's say, a proxy of handset volume -- unit volumes -- so we have a number of good design wins in TBS. And also, there's a little bit of a geopolitical tailwind that we believe is sustainable over a number of quarters. That's leading to our better-than-seasonal guide for Q1.

Great. Nice job on the quarter.

Thank you. I would now like to hand the call back over to Mitch Haws for any closing remarks.

That concludes today's call. Thanks to all of you for joining us today. We look forward to seeing you at various investor events over the coming weeks and at OFC starting tomorrow. Thank you.

Thank you again for your participation. You may now disconnect your lines.

Welcome back, everybody. I'm Joe Moore, Morgan Stanley Semiconductor Research. Very happy to have the executive team of Semtech, Hong Hou and Mark Lin with apparently some breaking news I'm told around a facility acquisition. Maybe you could give us an overview of what you just announced in the last few minutes.

Sure. Thank you, Joe. Good afternoon, everyone. You probably haven't get a chance to read the news yet. It's just crossed the wire. So we acquired a semiconductor laser company, indium phosphide-based lasers designed for high efficiency for optical transceivers and more ideally in the CPO, MPO modules and also for the tunable lasers for metro and DCI applications. So this leverage more than four decades of the innovation from the good old for us being around for a long time. Ortel was founded in 1980. It was the first company to commercialize the DFB lasers and Ortel was bought by Lucent. Lucent was bought -- that division was bought by Amkor when I was the CEO there. So now going a full circle, we just acquired that company called [ Hifil ].

Okay. And maybe help us what does that unlock for you guys? Obviously, you have a very strong business in sort of copper capable technologies and this push into optical. What does it help?

Yes. So definitely, we have been focusing on growing our business in the core asset area. Data center is one of them. And for optical transceiver space, Semtech is known to be a provider of the best TIAs for high-bandwidth optical transceivers with the data rate creeping up from 100 gig per lane to 200 gig to 400 gig and also the pursuit for low-power solutions, the LPO and also the DSPs with the manufacturer -- the advanced nodes that require driver solutions as well. So we are expanding our play -- our content in optical transceivers with the drivers. But that's about it for the electrical space. On the optical side, this really set a strong pillar for us to build a platform to provide comprehensive component solutions with the lasers and then we can get into the modulators and in the future, even photo detectors. But the lasers technology we bought in the fab we have indium phosphide-based high efficiency and provide wall plug efficiency well over 40%. So that's about the best, most efficient lasers in the industry. So certainly, that gives us an anchor to really build this comprehensive portfolio for 3.2T optical transceivers and CPO and NPO where they also need high-power lasers.

That's great. Thank you. So I always appreciate breaking news at our conference. Maybe we could wrap that into a discussion about your data center business. Can you talk about your overall positioning within data center and give us an overview of some of the products? And then feel free to wrap in anything that you just acquired into that.

Yes. So our data center -- we have a pretty diverse portfolio and our TIA and Tri-Edge business and TIA in the FiberEdge category. As we are reading the same report, there's going to be a very strong increasing demand for 800-gig optical transceivers. For single mode, we have a pretty good market share, about 50%. So that demand is growing, and that's the baseline. LPO introduction, we announced in the last earnings call that is going to be a milestone for the industry to move to high power efficiency, low-power consumption solutions like LPO and that start ramping.

So adding a growth driver on it. The importance of LPO also that's a prelude in my view, is MPO or CPO because it's all going to be DSP less solution. And then we have the copper product, as you alluded early on, the copper edge to provide redrive solutions to extend the reach in the cable or in the copper trace on the PCBs.

So the ramp for hyperscalers is going to start midyear this year. That's adding another step function. Then 1.6T, the early deployment, we were -- we had a very minimal play in there. but the industry is moving from 5-nanometer DSP to 3 to 2-nanometer DSPs in the second design cycle, we are front and center in there. So that's representing a step function on top of this baseline and a couple of incremental growth drivers. Now with the high-power lasers, we'll be able to provide optical solution co-optimized with our drivers and TIAs. So it's going to be good for the industry.

Okay. Very helpful. Looking at the TIA market, we've seen really strong market for optical transceivers -- there was other breaking news at this conference about prepurchases of optical components. Where do you stand within that supply chain? Are you tight? Is there other components that are tight around that? And that shortage of optical transceiver seems to be not quite at DRAM HDD levels of seriousness, but approaching that?

It's pretty tight. So we anticipated this because we have been engaging with our customers' customers very closely, like cloud service providers and GPU and CPU companies and ASIC designers as well. We know this is coming. And in my last earnings call three months ago, I was one of the priorities is to make capacity available. The current manufacturing partners, they are in allocation mode for the capacity. The good thing is we build ahead a little bit, and we -- that created a little briefing room and buffer that give us time to qualify additional foundry capabilities and capacity. So we'll be out right for the next 12, 18 months as we see the trajectory in our forecast.

Okay. And I guess, you're getting visibility into the next 18 months given the repurchasing activity and things like that.

That's right. So we have this long called a co-planning process with the lead customers. Their supply chain has the keen interest to make sure that their suppliers -- their supply chain is robust. So they give us forecast, we're able to get the wafers started way before their demand, but we have certain level of commitment, mutual commitment. And that model has worked out very well. We have expanded the co-planning process from a year ago, one customer to currently seven customers, so.

Okay. And is that growth also strong in 800 and 1.6, both .

Yes. So 800, we had visibility early on. So we were more prepared. 1.6T is coming, coming very strong, and that's a part we have additional capacity in the fab partner foundry qualified.

Okay. But that growth comes on top of what's still growing 800.

Exactly. Yes.

Okay. Helpful. Some of the new opportunities you talked about, copper edge and ACCs. Can you talk about the advantages of ACCs and where you're seeing initial traction there?

Yes. The ACC compared to the counterpart, the DAC, the direct touch copper, it can only go up to about 1 meter long. That's amazing. Actually, if you think about it. There is a 200-gig per lane, but cannot go longer than that. So the alternative is ACC or AEC. We know AEC has been very successful, demonstrated, but ACC consume 90% less power. And that is really very compelling to this hyperscaler, we worked with them a few months ago.

It takes a lot of effort for them to qualify for platforms. So now we are finally about ready to ramp on supporting their multiple applications primarily used in the backplane. The copper edge, in the meantime, it doesn't have to be integrated in the ACC cables. It can be integrated on the board to redrive to stretch the rich of the high-speed signal across the board and between boards. So that's -- I'm just so excited with every time when I talk to my product marketing guys. About a year ago, when we were start doing evangelical work and people were very skeptical and we turned them to somewhat curious. Now it's fully engaged. I will say by this time next year, when we talk again, I would definitely project the linear equalizer copper edge onboard is going to be a pretty substantial part of the revenue contribution.

Okay. So you have one kind of initial customer, but you think -- by this time next year, we'll be talking about more than one.

One initial customer on the cable for high volume. We got 3, 4 smaller volume customers -- but the linear equalizer onboard, we got multiple major customers ongoing right now.

Okay. And did that business participate in the NVIDIA ecosystem? Or is it mostly an ASIC.

It can be. It can be. Yes, the ASICs and GPUs and even the switches, wherever it is fundamentally, when you have a link budget issue challenges, say, from bump to bump, either 35 dB or [ 30 dB ] and every twist and turns, you have losses. So how are you going to be bridge the link budget? Well, you can use redriver or retimer. It's obvious redriver is just a very seamless and consume a lot less power than redriver. So I think that's based on that fundamental is going to be used in many different platforms by many customers and the industry is getting more educated. They feel more increasingly comfortable with the redriver solution by closely working with us starting from modeling to test to qualification corner test and to eventually volume deployment.

Okay. And how durable is that opportunity? I mean, it seems like -- some of that may migrate to optical over time as well.

Yes. So recently, I heard something like you say, "Oh, we need to put a terminal value on copper because this is going to be all optics. I can assure you that was my thinking about 20 years ago. Try to kill copper but the copper is getting stronger and stronger. So for scale up, it's always going to be -- copper is the predominant media. You have a printed circuit board. I don't know printed fiber optics board. So in the shorter distance, you are going to be using copper. This transition from copper to optics you pay overhead. So I would say probably up to 800 gig, you don't need optics to scale up.

But the co-packaged optics CPO is totally legit. They co-package with a switch die. So now the industry has a comprehensive ecosystem that can support liquid cooling and with one rack can consume 600 kilowatt or higher. And they can put the switch, the first level switch in the fabric in the rack, and that's co-packaged optics is first scale out, right? So scale up in special use cases, it may be happening in 2, 3 years. But right now, I think...

Maybe longer than that even. I mean, it seems like it a lot of the decisions around that have not been finalized yet, who's using what scale up?

Right. So the good news is we have solutions for copper, we have solutions for fiber as well. If they move from copper scale up to fiber scale up, fine. We have TIAs, we have drivers. Now we have lasers as well. With lasers, we can build silicon photonics modulator because our laser has the best far field pattern. That -- what does that mean is our laser light can couple into silicon photonics waveguide with very minimal loss.

Okay. Great. And then LPO, you've talked about in your initial comments, where are we with regards to potential adoption?

So the LPO adoption is a little different compared to ACC. ACC is more tied to a new platform design. So if they decide to use ACC and combine with the deck, and that's how all the backplane is going to be constructed that way. The LPO is more like the customers has been used it opportunistically. For example, in their system design, they all of a sudden found "Oh, there 2, 3 kilowatts shy in power budget." And they can say, "Hey, look, this is the same form factor, totally interchangeable." And for shorter reach, they can replace retimed transceivers with LPO transceivers. So we're seeing the ramps from the quarter we just finished, and we see this is going to be continued, but it's a gradual ramp.

The significance of LPO is really building the confidence in the industry is evolving into this MPO near package optics solution. You don't have to commit pin out and other perimeters for XPU for co-package, but leave the socket open and bring all the optics near the XPU. Then when your XPU ready, it just drop in, of course, you need to clear definitions through multisource agreement, whether is the geometry, pin out, keep out zone, shoreline density, all of that. We're very active in participating in that MSA, but NPO is stemmed from the good success of LPO in terms of signal integrity.

Okay. Great. So you have a lot of growth drivers in data center.

TIAs, I'm very excited about...

Like 6 months ago, we were talking more about new products, but the TIA market also looks very exciting, exactly, at this point.

And then after 200 gig, we have 400 gig, right? So the industry is continuing to push for the innovation in the upcoming OFC in the next 2, 3 weeks. We're going to showcase multiple chips from TIAs to drivers, to linear equalizers operating at 200 gig per second.

Okay. Great. Turning to industrial. LoRa was pretty strong in Q3. Can you talk about the outlook for that business and help us sort of size the exposure?

Yes. So LoRa continue to be very strong. Traditionally, the LoRa has been used widely used in three vertical markets. Smart metering for water meter gas meter is really long range, low power and very robust connectivity serves very well. That continues to be strong, but you only have so many meters and then the battery works for 15 years.

We'll continue to have a good business on that asset tracking and connected spaces. The most recent application drone has been really fantastic. The commercial drone and our product manager show the display, the Chinese New Year, they don't light up the fireworks anymore. They launched drone show each fleet, they get 20,000 drone that can change the pattern and other different colors, it's a fantastic -- and of course, the slow attitude economy in China, that's a very fertile ground for a lot of innovations.

So drone is fourth vertical -- and because of the bandwidth increase using dual band, we're not only using LoRa to send control signals, but they can transport with a very respectable bandwidth right now is at a 1.6 megabit per second.

That's increased from 10, 15 kilobit per second. So combined with some compression technology, people are developing use cases for edge AI. So they can take a snapshot, steel picture, send it back for archiving -- and for shorter distance like within a couple of hundred meters, you can even have the live video we demonstrated at -- at CES a month ago.

So that's high bandwidth, we benefited from that in unlock other vertical with edge AI. Then this Gen 4, LoRa plus, its LoRa plus another protocol, unlocked some other applications like security in smart homes. So we'll continue to expand the application use cases in different market verticals. We can easily see that business continue to grow at 15% to 20% year-over-year growth rate.

Okay. And can you give us a little bit more color on the Gen 4 multiprotocol, what that does?

So Gen 4, basically, people like the LoRa and -- but they wanted to have some other like a BLE protocol to say, the building managers, they wanted to send out kind of like a command to lock out all the doors in the building, and that works really well. But it has a little bit latency. And -- but you combine with another protocol like a BLE in there, near term is short is BLE and -- but not losing the LoRa feature. You basically have 2 IoT protocols in 1 -- and we have LoRa plus BLE, ZigBee, Thread, Matter and Z-Wave. And of course, each protocol, you need a software development kit to activate it, and we are doing one after another.

Like many silicon labs in a way.

Yes.

Okay. Great. And then maybe just to round out the markets, high-end consumer a lot of anxiety about memory constraints and what that might mean for the -- for some of the consumer businesses. Can you talk about that?

Sure. Mark you want to...

We're still pleased with our high-end consumer end market. In this area, we've been outpacing, let's say, just handset unit growth, and this is really due to Semtech's ability to capture market share and increase our content within the device. The last quarter, we talked about various TVS Transient Voltage Suppression devices. These aren't commodity devices. We're really protecting high-speed ports and to be able to clamp down the voltage and meet our customers' requirements on a high-speed data line.

It's not something easy to do. So in that case, we still feel very strongly about that market. We also have within that high-end consumer market is our per se sensing portfolio. Within that portfolio, we have what we believe is a leading-edge product there for sensing. Traditionally, it's been used in handsets to detect how close the handset is to a person's body, kind of lower the power, be form a way, meet specific absorption rate requirements. This is also an area where we had a nice small tuck-in acquisition at the beginning of the quarter. We're adding not only our per se product, we're adding a force sensing product, a force touch product. It's a really nice combination of 2 areas.

And this force sensing product that was really on our product road map. This was a clear way of maybe expanding on the make -- buy versus make thesis rather than wait 3 years, we had a portfolio that was available to us at quite a reasonable price. I think last quarter, we said it's well under 1 quarter of operating cash flow.

Okay. That was from Qorvo.

Correct.

Any other acquisition targets that you guys think about or any capabilities that you might need within really any of the businesses, which you know...

Yes. So we're constantly looking at our road map and most of the ideas and projects and desire stem from the interaction with the customers. when we provide a solution, they ask, do you have that? What about this? Okay. That's defined our road map. So Mark mentioned about this force sensor. It was on our road map and you come to make or buy opportunistically when that asset became available, we just snapped it. Here, this acquisition we just did the HFO in that category as well, there has been a little bit longer pursuit.

We know that for the data center growth in order to grow the content, for example, we have the aspiration for 800-gig optical transceivers. We provide TIAs. Our content is only high single digit. I wanted to be growing into high double digits by 3.2T. So we got to grow, but you can't really do the me-too to chase the market, especially when the market grows so rapidly, and we have to be able to think ahead and build some differentiations.

So that's on our road map, and we pursued others and didn't come through, but this one, we drop to the finish line. This is a very exciting opportunity for us to put another anchor to build a platform. I use that analogy. As for other targets, certainly, we are having strong aspiration. We got more ideas to drill, but nothing at this point. It's close to...

Sure. I think we've been pretty prudent and selective in targeting on acquisitions. And we're bringing that same muscle memory from how we invest in our organic R&D. R&D is -- we increased R&D, but the return on this investment for our R&D, it's been proving out LoRa data center per se. And this acquisition that we just announced today, consideration is about $34 million. So very, very prudent use of capital to really expand Semtech's market share, markets.

Yes. That's great. Maybe we could talk about the Sierra Wireless situation. How do you think about the outlook for IoT Cellular? And is the divestiture of noncore assets still something that you?

Yes. Maybe I'll address the divestiture. So this is of the cellular module business, which we've spoken about and we discussed on our last call. This business is still performing quite well. Backlog and order rates are still quite strong. This business though has about a 20% gross margin. So it just doesn't fit in Semtech's portfolio. It's a great business that fits in someone else's portfolio.

The process for divestiture is ongoing. It's an active process, and we have multiple touch points a week and diligence calls and whatnot. So it's difficult for me to peg a time in terms -- because it moves the deal speed, but I'm very -- I believe that the deal is pending, optimistic that we can get this to a close.

Okay. And then what do you think the gross margin profile of the company looks like post divestiture.

Sure. Just historically, if you strip out this business, we're approaching close with striking distance of 60% gross margins. And what Hong was talking about in these growth areas, especially in data center, right, it's accretive to that -- very accretive to that 60% gross margin. Last quarter, we reported that our Signal Integrity Products segment had a 65% gross margin and new product introductions would be accretive to that 65%. So really kind of for a company in our space, having a gross margin starting with a 6% is really where we should be.

Yes. Great. Well, look, you've had a great tenure as CEO. Maybe just I'll open up for questions, but just to close for me. The opportunity that you've delevered the company and now you're able to focus on these growth drivers, it seems apparent that sort of that delevering process was a major focus in the beginning. At this point, you're really focused all forward on growth.

That's right. Yes. So we have come a long way in a short time. So certainly, when I started about 20 months ago, with the CEO. So we kind of map out the 3 stages. It's a little bit tongue and cheek. We say some tech rising. We need to absolutely beef up our balance sheet to fix the balance sheet. We had a leverage ratio of 9.5x, $1.3 billion net debt, and we do not have any operational freedom to do anything aggressive.

So Mark has led the effort. We benefited from the good market as well. We did the financing. We now today totally fixed the balance sheet. We have well over $600 million liquidity. We can do things somewhat more aggressive. But even in that time, the second pillar of the Semtech Rising is we are a technology company. We have to delineate what is the core, noncore and invest -- accelerate invest in the core area.

So that's why we were able to gain market share in the data center. That's why we were able to expand our portfolio with a limited financial capability. And the third thing is the employee morale. We're not sure what the future of the company is through the Semtech Rising initiative, we built the transparency, build the trust and getting the whole company going into the same direction.

So that has been great. Then what? The second step was the Semtech transforming. So it's again, tongue and cheek, but it's really significant. We wanted to optimize our portfolio, some noncore assets we're doing divestiture. I would say we're well within 20-yard line.

So -- and I'm optimistic it's going to happen. And then we're adding things in the core asset area and to really evolving ourselves from the commodity component second source supplier to the leading solution provider. We wanted to be in a room when a new standard, new solutions come about.

For example, last week, we instigated and founded the MSA for ACC that drive a standard and benchmark how to do test, what kind of performance you should expect. We're going to do the same thing on MTO side. And then we have, again, as I said, have a pretty aggressive plan and very exciting road map. We're going to increase R&D investment. Now we have the capability, we can do that. And we're going to continue to look around some tuck-in opportunities. Then the third step, we wanted to be the industry leader in the area we are in, Semtech excelling, deliver higher growth rate than peers, deliver higher operating margin than peers. So we -- I think we're well underway and we're pretty excited about what's ahead of us.

That's helpful. I actually did have one additional question. Just your relationship with hyperscalers, that seems really important when you're sort of pushing in these data center markets. Are they coming to you with projects, things that they want, sharing their road map with you?

Yes. So it's not easy. Sometimes it just take the persistence, but the overarching principle is the customers, they are so busy. They don't want to meet you if you don't add value. So good thing is in the beginning, I had a good connection and I was able to keep the doors open. And when I got my guys in, they really see the analog design is not easy. We are really known, that gentleman in the back was telling me that I was so happy to hear that. Yes, so we were able to do modeling end-to-end signal integrity in the beginning, you must have a model, we have a model as well. Now they realize Semtech has a really very robust model. So before they define a platform architecture, they wanted to check what are the boundary conditions. Can this be achieved or not? So we're the guy, and that's really a great position to be.

Great. With that, let me see if there's questions from the audience. If not, we'll wrap up there. Thank you so much.

Thank you, Joe.

Thank you Joe.

Okay. We're going to get started. I think this might be the last session of the day, but we're into late afternoon. So very pleased to have Semtech. I'm Tim Arcuri. I'm the semi and semi equipment analyst. Pleased to have Semtech. We have Hong Hou, who is the CEO, and we have Mark Lin, who's the CFO. So thanks, Hong and Mark.

Thank you.

Thank you.

So let's just -- I just wanted to go back, and I thought your quarter was sort of a watershed quarter in a way because now you've got a large customer ramping ACC and really proving out the design. So can you just talk about that? Can you talk about sort of review some of the highlights of the quarter and maybe talk about the data center opportunity, really?

Yes. Thank you. So we just announced our Q3 results last week, early last week, and we achieved the sequential revenue growth. And in our core assets, I want to highlight in the data center area, we had the revenue growth sequentially 8% and we project the revenue to continue to grow in an accelerated pace in Q4 for 10%. We also announced -- gave a detailed schedule for ACC ramp. Our hyperscaler customer gave us the time line to start ramping incorporate ACC in their rack by mid-2026. And then we supply our linear equalizer IC to the cable manufacturers, they need 3, 4 months to get our IC integrated into the cables. So we will start seeing the ramp in April quarter for ACC.

And what is this -- I mean this customer tends to set the tone, I would say, we'll say, for other hyperscalers. So what does it mean? I mean, sort of what does this confirmation mean for how quickly ACC can penetrate other customers?

Yes. So this hyperscaler has a tradition to lead the industry with the disruptive solutions. When they started using 200 gigabit per lane service, so they certainly look at a different connectivity solutions than we have been engaging with them maybe 10 months ago. They started looking at a deck, ACC, then AEC, their primary use cases is interconnect, the different XPUs within the rack and then between racks. So they feel that for the data rate, they wanted to transport -- transmit 1.6 gigabit per second over 6 lanes -- 8 lanes, they can achieve everything they need with a 3-meter reach with the ACC. Below 1 meter, from 0 to 1 meter, they can use 30-gauge deck cable. So now they down selected this between DAC and ACC. And the clear partition below 1 meter is DAC and above 1 meter to 3 meters is ACC.

And yes, I guess, just thinking about how this could catalyze other hyperscalers.

That's right. So they have been the leader in the market in the hyperscaler -- among the hyperscalers. They always pioneered the new technology. I am sure there's going to be a catalyst for other hyperscalers to use ACC in their fabric for direct design and also even the scale-out.

And can you just talk about what your, sort of, I don't like to use the word right to win, but what's your edge with copper edge? Is there room for you to expand content beyond ACC, like maybe sockets in the backplane, for example.

Yes. So first of all, for ACC, the reason they chose ACC versus AEC, clearly, it's a power consideration and the latency is another consideration. When you have ACC can do what AEC can do as well within 3 meters, but with a 90% power saving, they will jump on it and also the latency is almost negligible with ACC. Beyond the cable, you asked about the linear equalizer can be integrated on the board to extend to reach of high-speed links as well. So we have several use cases in the evaluation and qualification stage with other customers.

And does it matter -- I mean is it agnostic for ASIC versus GPU?

It doesn't matter. It depends on the quality of the SerDes and for now, the 200 gig per lane, there's really 2 -- only 2 primary suppliers for 200 gig SerDes. So we work with both of them pretty closely, and we know the characteristics and our product has been tuned to compensate their SerDes characteristic to extend the reach.

But there's no pricing difference. There's no content difference.

Oh, I see. I see. So then our linear equalizers can be integrated in a paddle board of the cable -- ACC cables. The same design, same content can be surface mounted onto the PCB trees then to extend the reach. So the content is about the same. And the technology is identical. It's just the form factor making a little change in order to accommodate easy integration on the board.

Great. Let's talk about -- we talked about ACC, let's talk about LPO. LPO is a little more plug-and-play than ACC has been. I mean, now you've sort of crossed the chasm, I would say, in ACC, but LPO is more plug-and-play. So can you talk about what the opportunity is there and how to size the LPO opportunity relative to the ACC opportunity?

Yes. So you're right. The LPO is more plug-and-play because it's the same form factor with a DSP-based retime solution, optical transceivers, same electrical plug-in interface, same optical plug-in interface, same mechanical form factor. As long as the customers do the test and to see the signal integrity is maintained between their links. For example, the first layer of the switch fabric from server to top of the rack, they can optimistically use low power consumption of the LPO and it's also low latency, a little lower price than the DSP-based solutions as well.

So we see more broad acceptance of the LPO with reference to 3, 4 years ago, the industry starts talking about that, there has been some resentment, but now I see that the entire industry really getting a broad acceptance of this technology. So we see some acceptance, and we see some adoption in some first stage of the switch fabric. So you will see the gradual ramp. And in Q4, we're going to be having the first quarter milestone to see the primarily TIA for the LPO at this point reached a mid-single-digit revenue level.

And have the interoperability concerns for LPO largely been resolved? I mean. I think -- and part of that is LPO broader than just Broadcom-based systems?

Yes. So I think the LPO, one thing -- one key enabler is a clean signal noise ratio from the SerDes. The Broadcom system is really -- is excellent. Tomahawk 6 ports, the signal integrity is great that enable the adoption of LPO.

So can you size for us maybe just go through some of the relative sizing of the ACC opportunity versus the LPO opportunity?

So the LPO is going to be ramping gradually. So it's going to be, in a way, cannibalizing the DSP-based retime solutions. It's ramping gradually. So I would say probably if you say 2, 3 years later, reach a steady state, maybe overall 800-gig transceivers, pluggable transceivers, you can expect 25%, 30% of the mix is going to be LPO based. Transition to LPO is beneficial to Semtech because we're going to be able to expand our SAM, not only providing TIAs but has the opportunity to provide the drivers. But size-wise, I think ACC from now, we look at it, it's a much bigger opportunity than LPO for us.

ACC is bigger. Yes. And can you just -- just to wrap up on LPO, can you talk about your opportunity and how it changes between 800G and 1.6T?

Yes. The LPO for 800 gig, it's, I would say, pretty mature. For 1.6T on the other hand, the industry is still working on the signal integrity or reflection issue from the transmitting path. Receiving path is good. So some customers wanted to capture the low power benefits of that by rolling out LRO for 1.6T first. So we see the near-term 800 gig, no one cares about LRO and it goes directly to LPO. But for 1.6T, it's going to be going through LRO step, but I do believe eventually the industry will figure out a way to get LPO available for 1.6T.

So you -- but your opportunity is probably bigger in 800G, is that fair?

So the opportunity for 800 gig, we definitely -- our TIA has been broadly widely regarded as the best solution by the industry. We have the lion's share for the TIA side. But driver was a little late, we're playing catch-up and we have some customers designing in use our 800-gig LPO drivers, which we do believe the only driver have to comply with MSA, multisource agreement and provide, say, for example, automated gain control and digital diagnosis capabilities.

For 1.6T, we do not want it to be late. That's why we're accelerating our road map. We'll make the LPO compliant 1.6T LPO driver available before the end of the year, so we can sample to our customer base to have them play around, give us feedback. In case if we need a respin, we'll have the base for it.

And is there anything -- any loading commentary you could make for 2026. I mean you haven't guided the year, obviously, but can you give us any like puts and takes in terms of loading commentary, how to think about. I mean, you've got the -- you've the consumer business, too. We haven't talked about that. But give us any loading commentary for like how to think about how the year could play out next year?

Yes. So for the data centers, it's just like we talked about the LPO, we talked about ACC. And another thing is 1.6T transceiver ramp. We are capturing that opportunity as well. Even though a lot of volume for 2026 calendar year is going to be 800 gig optical transceivers. For U.S. hyperscalers, the projection is the 50 million units. That's a significant number compared to, say, calendar year '25, probably will be ending up like 30 million units.

So we have a solid base and then overlay the 3 growth drivers on top of it. LPO transition bring more content. SAM is 250% of the DSP-based transceivers. ACC is a net revenue adder. And with this hyperscaler, we project a pretty rapid ramp in the second half of the year than the 1.6T. And the other segment, I will hand it over to Mark to talk about it. We got multiple drivers as well.

Yes. So I'm pleased that say Semtech is firing on all cylinders across all of our end markets. So I'm pleased in that area. High-end consumer, we've released our results for Q3. We're clearly outgrowing, let's say, handset volume as a proxy, right? So that is a great indication where we're gaining market share. We're getting content on devices. in handsets and in wearables, right? That's a new growth area for Semtech.

In the industrial side of the business, that business is also inflected to growth. LoRa, we posted over a $40 million quarter in Q3. In that area, we have multiple growth drivers as well. This is where we put in some good amount of investment in the past, and that's really resulting in some very nice top line growth for LoRa. LoRa, we can see growth in dual band, so sub-gigahertz, which is what LoRa is known for, plus 2.4 gigahertz that's expanding a lot of market share, being able to transmit a photo, an image, is really increased adoption, let's say, in the drone market.

And then another area is LoRa Plus, another area where we can introduce LoRa Plus another protocol like Z-Wave, really expanding the use cases for LoRa. We received some very nice growth. As we said on the last call, we're ranging from $30 million to $40 million as a base per quarter, growing at about 15% to 20% CAGR.

And can you talk about the force sensing business, the new force sensing business?

Yes. So we identify the 3 key areas, the core asset, data center, LoRa for IoT and then sensing. So we have our portfolio of the technology in a product, the person sensing for a specific absorption read and also [ gesture control ]. That's based on capacitive sensing capabilities. And when we did the road map exercise with our technologies, we always feel that adding force sensing onto our existing capacitive sensing and [indiscernible] sensing technology will broaden our capability, will allow us to penetrate in a broader customer base. It just so happened this technology, this asset available we grabbed it and this is a really pretty exciting opportunity for us.

We've got this -- acquired this asset from Qorvo. They acquired a company called NextInput 4 years ago. They have about 175 issued and pending patents. We -- together with the deal transaction, about 50 key employees came with it. So we are already addressing some common customers with the combined solutions. We do expect this combination by leveraging our global sales and supporting staff will help us to accelerate the force sensing business, but also pull through our capacitive sensing business. So this is very synergistic.

So is there any way to size like how much the synergies could be the force sensing business combined with pulling through your other businesses? Like is it crazy to think that next year, I would think it's probably not going to be that much. But in 2027, is it -- I mean could it be a $40 million or $50 million type of an outcome having acquired that.

That's consistent with our model and thinking and projection.

Okay. And then can you talk a little bit about just about the asset divestitures that's something that you've been fairly consistent that you want to clean up the product portfolio more. So can you just talk about that?

Yes. We've been pretty consistent among the portfolio. When we had a balance sheet limitation, I will say that way, but we still identify some growth opportunities. We delineated the core assets. For those assets, we increased R&D investment over the past 18 months. But for some noncore assets like the cellular module, specifically, to us, it's a challenging. We got a great team and it has been integrating well and has inflected growth, but the margin profile is very different from the semiconductor from even the other part of the Sierra Wireless acquired asset.

So it's low margin to us, but to the right acquirer, it's the great margin to them. So this time around, we hired different adviser for [ disclosures ]. So UBS is a transaction adviser for this time, we map out the target space and think about the synergy and did very targeted marketing. And we received multiple indications. We're in the process of due diligence. So that's -- with that part carved out, we'll be able to clear up a lot of noise and solve this margin disparity issue.

And then for the employees, the business going with the transaction, they will have a good home, and they will be viewed as a core asset in the new home. So I think it will be good for that business to continue to drill and serve the broader customer base.

And following the divestiture, we're looking at gross margins approaching 60% starting with the 6%. That in and of itself is a really compelling reason for the divestiture. And it's really going on top of our Signal Integrity Products business, which in Q3, we reported a 65.1% gross margin. That segment includes all of our data center business, right? So if you're looking for where we consider growth areas, data center plus ACC plus LPO plus the transition 1.6T. This is a very clear path to 60% plus gross margins for Semtech on a consolidated basis.

And Mark, do you think -- on kind of a pro forma basis, do you think that you'd be able to make it earnings accretive?

Yes, we do.

You do, okay. And just in terms of sort of other optimizations of the portfolio and like how you can -- how synergistic are the -- if we pro forma the business, how synergistic from a development perspective are the remaining businesses going to be?

Yes. Let me just say, at this point, we're really focused on this first transaction, right, the cellular module transaction. We believe that's going to be transformational on gross margin on a consolidated basis. And then after we get that project completed, we'll be looking to maybe provide some other insight into further thinking.

Can we talk about LoRa? So you did say that LoRa is between $30 million and $40 million is your thinking in sort of the near term. I think you guided it back to the middle of the range for December quarter. So can you just talk about the drivers there? I know Gen 4 is gaining a lot of momentum. But just can you walk through that business a little bit for us?

Yes. So LoRa has been traditionally used in 3 vertical -- market verticals, smart metering for the water meter and gas meter, connected spaces for the industrial automation and asset tracking because of the broad coverage and low power. So that's a good sweet spot. And because of the dual band increased the bandwidth from 15 kilobit per second to 2.5 megabits per second, that's significantly increased, that's unlocked some opportunities in the commercial space. For example, commercial drone using aerial survey, you'll be able to snap pictures and transmit back the [ steel ] pictures then giving some -- adding some functionality. And then this LoRa Plus with other RF protocols unlock opportunities like security, smart building. So we are adding more application verticals, and that's the basic driver for growth of the business.

And Mark, can you talk about -- so as you look into the April quarter, what sort of -- I'm not asking you to guide April, but what's the sort of normal seasonality? I know that the IoT business has a lot of seasonality in April. But how to think about the puts and takes seasonally?

Yes. Really, in the April quarter, the only seasonality we somewhat see is in that industrial business, but it's really related to one portion of the business, kind of the routers and gateways business. Other than that, seasonality is typically in the fourth quarter, right? So when we're in high-end consumer, right? Typically, it's down 15% to 20% historically, just based on let's say, smartphone builds. We guided our Q4 down in the low single digits, though. So that's another, call it, proof point of market share gain.

And then this is just more of a -- just how to think about the dilution. There's been a lot of volatility from the converts. So can you just talk about that and just how many shares do we have?

It's absolutely something that we want to address in terms of dilution. So without getting to the math, if you look at our earnings presentation that we issued along with the last call, we have a dilution table. But needless to say, that's something that we want to address. We had 2 converts outstanding. Just last quarter, we took out one of them completely. The other one we reduced by 1/3. So on the 2027 converts, it was about $320 million outstanding. We took that down to about $101 million. Then we issued $402 million in new convertible notes, 0% coupon, and we have a cap call up to 100%. So really, it becomes dilutive at north of around, let's say, about $142.

So we've taken care of dilution, we believe, and there's just going to be a lot less volatility due to converts. I think those are both beneficial.

Got it. And then can you walk through just -- this is -- I mean, let's assume pro forma of the business. Can you walk us through some of the puts and takes as you think about gross margin as we go through '26. Can you walk through some of the relative margins of those remaining businesses?

Sure. I'd say it's mostly to the positive. I think there's, in my view, a little bit more upside opportunity than downside risk. So like we said, Q3 gross margin for Signal Integrity, a reported quarter. We had 65.1% gross margin. ACC is additive to that. LPO is additive to that. Other areas, just total portion -- total part of the business, LoRa being an asset that can grow pretty nicely as gross margin starting in the 70s, right? So north of 70% gross margin, that's pretty healthy. So in terms of overall, like I said, if we were to divest the lowest margin portion of our business, again, low margin for us, great margins for somebody else. We quickly approach 60% gross margins. In the growing areas of business, we can surpass 60% in the near term just based on our expectations of growth over the next fiscal year.

And Hong, let me ask you about China. So what are your opportunities in China with the Chinese CSPs, as they start to build out racks?

Yes. So China is an important market for us. We ship our ICs going through distributors in Hong Kong. Some are for the China domestic use, and a lot of them are integrated in China or Southeast Asia, then imported back into the U.S. So in the U.S., we got all the hyperscalers accelerate the data center build-out for AI applications. For China CSPs because of the availability limitation to -- of the GPU, the most advanced GPUs, so their acceleration rate is not as high. And also, most of the optical transceivers, for example, is 100 gig per lane, 800 gig level, but it's still pretty healthy in growth.

So we have a very broad penetration. We probably, as of September -- well, a couple of months ago, we were able to address the last customer that we had no component in it. So we supply to everyone, every optical module suppliers. Some of them are more focused on China market, and some of them are mostly focused on the U.S. market. So they are not growing as fast as the U.S. market, but still demonstrated sequential growth.

Great. And then last thing, Hong, I wanted to ask you, you mentioned on the call, you said your 2 priorities. One is to your growth opportunities in -- with your core assets; two is to divest the noncore assets. But then you said to fill capability gaps. So where do you see your capability gaps?

Yes. So we wanted to clearly -- the data center is a key area we wanted to grow today in our transceiver content. In the transceiver content, our contribution is still limited. It's just still limited at the physical media device level, drivers and TIAs. It will be great to have the optics. But we're looking at differentiating capabilities, not just for the sake of it. We can have the operational leverage, but we are looking for differentiated capabilities that we can co-sell with devices. So -- and we are looking at other areas as well. But for now, we wanted to focus on making the capacity available to capture the growth opportunity and never wanted to miss the opportunity because of the capacity because the lead time limitations. So we're addressing that proactively.

And these are things you have to acquire? Or do you think that you could...

This thing is more the operational focus, but then some gaps are internally developed because we have multiple generation road map alignment with some key customers. We know where the gaps are. So going forward, now our balance sheet is much stronger than before. We'll be more heavily indexed in R&D growth in the data center area, R&D investment in data center area to accelerate the growth in the future years.

Perfect. Well, we're out of time. Thank you, Hong and Mark.

Thank you so much, Tim.

Good day, and thank you for standing by. Welcome to Semtech Corporation's Third Quarter Fiscal Year 2026 Earnings Conference Call. [Operator Instructions] Please be advised that today's conference call is being recorded.

I would now like to hand the call over to Mitch Haws, Senior Vice President of Investor Relations for Semtech. Thank you. Please go ahead.

Thank you, and welcome to Semtech's Third Quarter 2026 Financial Results Conference Call. Participants on today's call are Hong Hou, our President and Chief Executive Officer; and Mark Lin, our Executive Vice President and Chief Financial Officer.

But before we begin, I would like to highlight upcoming investor events, including the UBS Technology Conference on December 2 and 3; the Consumer Electronics Show on January 6 through 9; and the Needham Growth Conference on January 13 through the 14.

Today after market close, we released our unaudited results for the third quarter of fiscal year 2026, which are posted along with an earnings call presentation to our Investor Relations website at investors.semtech.com.

Today's call will include various remarks about future expectations, plans and prospects, which comprise forward-looking statements. Please refer to today's press release and see Slide 2 of the earnings presentation as well as the Risk Factors section of our most recent annual report on Form 10-K for a number of risk factors that could cause our actual results and events to differ materially from those anticipated or projected on today's call. You should consider these risk factors in conjunction with our forward-looking statements.

We will refer primarily to non-GAAP financial measures during today's call. Please see today's press release and Slide 3 of the earnings presentation for important information regarding notes on our non-GAAP financial presentation. The press release and earnings presentation also include reconciliations of our GAAP and non-GAAP financial measures.

With that, I will turn the call over to Hong.

Thank you, Mitch. Good afternoon to all of you joining the call today. The Semtech team made solid progress again this quarter, driving strong sequential and year-over-year revenue and earnings growth, aligning our data center road map to capture major growth and design win opportunities ahead, further strengthening our financial profile, all while executing on the R&D road map and portfolio expansions that we believe establish a foundation for growth.

Looking at the Q3, net sales were $267 million, up 4% sequentially and up 13% year-over-year, driven by the momentum of our data center and the LoRa portfolio. Adjusted operating margins grew 180 basis points sequentially and 230 basis points year-over-year. Adjusted diluted earnings per share were $0.48, up 17% sequentially and 85% year-over-year.

Again, this quarter, the core assets we have delineated, namely data center, LoRa and PerSe together strongly contributed to our revenue growth. We continue leveraging our R&D resources to expand our portfolio, including in LoRa with multiple protocol integration showcasing Wi-SUN and LoRaWAN synergy for smart infrastructure and the new TIA and driver building blocks that establish a new performance standards for 1.6T multimode optical transceivers in AI data centers.

In addition to our strong financial performance, we further optimized our capital structure with a successful convertible offering. The collective actions taken over the past few quarters have provided Semtech significant balance sheet flexibility, resulting in nominal interest expenses and a much-improved cash flow generation. This improved financial position allows us to accelerate investments in our core technologies.

Finally, portfolio optimization remains a key focus. At the beginning of our fourth quarter, we completed the acquisition of the force sensing business, including its technology, products and key employees from Qorvo. By leveraging Semtech's customer penetration, global sales and support network and our existing capacitive sensing product portfolio, we expect to accelerate the proliferation of the advanced force sensing Human Machine Interface solutions and MEMS sensors by targeting leading computing, smartphone, wearable and automotive applications.

In addition, we are making solid progress on the divestiture of non-core assets. With our new financial adviser, we have engaged in diligent conversations with a number of interested parties, which has generated multiple indications of interest. We believe this assets represent a very compelling synergistic value to this potential acquirers.

Now, let me move the discussion to our end markets. For Q3, Infrastructure net sales were $77.9 million, up 6% sequentially and up 18% year-over-year, strongly supported by our data center business. Net sales for data center were a record $56.2 million, up 8% sequentially and up 30% year-over-year, benefiting from strong demand for our broad portfolio, including our market-leading FiberEdge TIAs whose net sales set another record.

Moving into Q4 and the next fiscal year, we expect an acceleration of sequential and year-over-year growth for our data center business. This conviction is supported by our expectation of continued increases in AI CapEx, expanding customer engagement and a strong demand pipeline for our high-performance low-power solutions, including the incremental contributions from Linear Pluggable Optics, our LPO and CopperEdge linear equalizers.

We believe our low-power analog solutions are a core enabler for making next-generation data center infrastructure scalable at 800 gig and 1.6T. With the hyperscale and AI data centers' capacity measured on electric consumption, every incremental watt saved in networking connectivity multiplied by tens of millions of ports will enable a meaningful increase in compute capacity.

By delivering best-in-class efficiency and signal integrity at the fiscal layer, analog solutions give cloud and AI operators the flexibility to adopt new 1.6T-based topologies, whether that is the higher density switches, new optics architectures or more disaggregated racks while staying in within strict power, thermal and transmission latency envelopes.

To support data center build-outs, we are seeing broad-based demand acceleration, supported by customer forecast for 800-gig TIAs through 2026. Beyond 800-gig, we are actively supporting a wide range of customers on their 1.6T transceiver designs and deployment with both TIAs and drivers, and we expect 1.6T volume ramps to begin early in calendar year 2026 and concurrent with the deployment of 1.6T switches.

Regarding LPO, we have secured design wins with several leading U.S. hyperscalers with our TIAs and drivers in 800-gig transceivers and AOCs. And we continue expanding our customer pipeline through engagement with our optical module customers. We expect a meaningful revenue contribution from TIAs for LPOs starting in Q4 and the momentum to build into calendar 2026. In parallel, we are accelerating our R&D road map and targeting initial sampling of 1.6T LPO drivers and TIAs before year-end.

Regarding active copper cables, customers benchmarking ACCs against the competing technologies are seeing clear advantages, excellent signal integrity, lower latency, and more importantly, power consumption up to 90% lower than DSP-based AEC solutions. We expect to ramp ACCs with a major hyperscaler during calendar year 2026. With this deployment transitioning incorporating ACCs in place of AEC or DACs, we anticipate broader market penetration as this hyperscaler demonstrates ACC's benefit versus incumbent technologies.

Our engagements with additional ACC customers are intensive and broad-based, and we anticipate more design wins over coming quarters. In addition, a number of customers, including our android customer, are evaluating the integration of our CopperEdge linear equalizers on their PCB boards and connectors to improve signal integrity of high-speed links. We anticipate design win of onboard CopperEdge use cases over the coming quarters.

Moving forward, we believe our broad portfolio of FiberEdge TIAs and our rapidly emerging CopperEdge and LPO solutions position us for accelerating data center revenue growth throughout 2026.

Now moving to our High-End Consumer end market. Net sales for Q3 were $41.9 million, up 2% sequentially and up 5% year-over-year. Year-to-date net sales were $118.5 million, up 6% compared to the same period last year. Growth from our High-End Consumer's portfolio is outpacing market metrics such as worldwide handset unit volume growth by a considerable margin, demonstrating market share gains, customer adoptions of our differentiated solutions and the strong supply chain execution.

In addition, our PerSe sensing technology continues to be designed in a growing range of applications, including smart glasses and smartphone platforms supporting both existing designs and new launches over the coming quarters. As I referenced earlier, we completed the acquisition of leading force-sensing portfolio from Qorvo at the beginning of Q4.

The integration is well underway with our first product shipped starting last week, and we look forward to this combination expanding our sensor portfolio with a proven IP and paired with our global go-to-market engine, unlocking cross-selling opportunities across a diverse array of leading customers. The combination of this unique capabilities provides a robust set of touch and gesture detection capabilities.

Moving to our Industrial end market. Q3 Industrial net sales were $147.2 million, up 3% sequentially and up 12% year-over-year, driven by another quarter of strong LoRa performance. LoRa-enabled solutions net sales were $40 million, up 10% sequentially and up 40% year-over-year, supported by the continued expansion across several end markets and multiple applications in verticals such as smart utilities, smart building, smart city and asset management.

Looking ahead, we believe we are well positioned to drive LoRa adoption with additional capabilities and features. Our recently launched Gen 4 LoRa Plus transceivers offer integrated multiprotocol connectivities in addition to the LoRaWAN capabilities in a single chip across both sub-gigahertz and 2.4 gigahertz frequency bands. This simplified hardware design lowers BOM cost and enables customers to create unified design, supporting multiple protocols, thus enabling deployments for customers rolling out solutions across different geographies and regions.

The LoRa Plus transceivers now deliver data rates of up to 2.6 megabits per second on both sub-gigahertz and 2.6 gigahertz band. This capability enables faster transfer of video, images and a richer sensor data while maintaining ultra-low power consumption and enables applications that were not practical before.

We're also continuing to see good traction in commercial drones. LoRa enables long-range communication, up to 10 kilometers for applications like agriculture monitoring, livestock tracking and infrastructure inspection. With Gen 4's higher data rate, drones can now transmit images and sensor data in real time while covering larger areas efficiently.

Our IoT systems and connectivity business recorded Q3 net sales of $88.3 million, down 1% sequentially and up 7% year-over-year. We see strong design win momentum as IoT transitions from 4G to 5G, leveraging our market leadership. As of this quarter, we have completed all the necessary certifications for our 5G RedCap modules, and the products are now commercially available.

The business pipeline continues to be strong, thanks to the broader market recovery and the favorable geopolitical environment for this business. Networking solutions with routers, gateways and Airlink services in the portfolio had a strong execution quarter, advancing strategic initiatives across carrier partnerships, software platform innovation and the market positioning. We expanded our 5G stand-alone capabilities with support for network slicing, enabling dedicated first responder network slices on T-Mobile's key priority and Verizon's frontline networks.

We believe this positions Airlink as a differentiated solution for mission-critical public safety communications where quality of service and network's prioritization are essential. We launched AI-powered support tools, delivered our next-generation management platform supporting both cloud and on-prem customer requirements and announced a strategic partnership with [ GTEC ], extending our reach by embedding Airlink connectivity into their rugged computing ecosystem.

The mission-critical cellular router market continues growing in double digits with accelerating 5G refresh cycles, and we believe we are well positioned to capture share through our carrier relationships, ecosystem partnerships and differentiated ruggedized solutions. We also launched the industrial first single vendor offering with Skylo, providing access to terrestrial and satellite networks through a single SIM and delivering the industry first complete device-to-cloud terrestrial and satellite IoT solution from a single partner.

Our strong results this quarter reflected the impact of our focus on growth of our core assets, disciplined R&D investments and the deep and expanding partnerships we are building with our customers. As power constraints intensify for our customers across all our end markets, we believe Semtech is uniquely positioned to lead with ultra-power efficient solutions spanning high-bandwidth data center networking, LoRa connectivity for rapidly expanding IoT use cases and sensing technologies that enable the functionality of next-generation AI interfaces. We see significant opportunities ahead and are focused on executing against them while continuing to create long-term value for all of our stakeholders.

Now, let me lay out my priorities for the next few months. First, capture growth opportunities in our core assets. Through selective strategic investments, we plan to fill key capability gaps. Leveraging our operational excellence, we will also focus on ensuring capacity availability, particularly against the backdrop of tight supply and geopolitical uncertainties.

Second, focus on the divestiture of non-core assets. This will help address margin disparities and enable us to focus fully on our core business priorities. Third, strengthen our winning culture and elevating our company mindset to work [ great ] is the new normal.

In the year of Semtech Rising, we fixed the balance sheet, aligned our core portfolio with market growth drivers and built a strong foundation of winning culture. Building on the momentum of these successes, we are now embarking on the journey of Semtech Transforming, paving the way towards Semtech excellent and solidify our position as a global leader in enabling next-generation data center, LoRa-based IoT and our expanded sensing portfolio.

With that, I will now turn the call over to Mark for additional details on our financial results and our outlook for the fourth quarter of FY '26.

Thank you, Hong. For Q3, we recorded our seventh consecutive quarter of net sales growth with record net sales of $267 million, above the midpoint of our outlook and up 13% year-over-year. Net sales trends by end market, reportable segment and geographic region are included in the accompanying earnings presentation.

Adjusted gross margin was 53.0%, at the midpoint of our outlook. Total semiconductor products gross margin was 61.3%, up sequentially from 60.7% and up year-over-year from 59.9%. Total semiconductor products gross margin reflects meaningful sequential and year-over-year net sales gains in data center and LoRa. IoT Systems and Connectivity gross margin was reflective of mix related to net sales growth in cellular modules with Q3 at 36.6% compared to 39.5% in Q2 and 41% in the prior year period.

Adjusted net operating expenses were $86.5 million, below the midpoint of our guidance range, benefiting from prudent cost control and a relatively stronger U.S. dollar. Adjusted operating income was $54.9 million, up 13% sequentially and up 26% year-over-year, resulting in an adjusted operating margin of 20.6%, up 180 basis points sequentially and up 230 basis points year-over-year.

Adjusted EBITDA was $62.7 million, up 11% sequentially and up 23% year-over-year. Adjusted EBITDA margin was 23.5%, up 160 basis points sequentially and up 190 basis points year-over-year. Adjusted net interest expense was $2.5 million, down 86% year-over-year.

The capital structure changes completed in October were in effect for only 2 weeks of the quarter, so the bulk of the benefit will be realized starting in Q4. To summarize these changes, we issued a $402.5 million convertible note, inclusive of the Greenshoe due November 2030 with a coupon of 0% and a conversion premium of 42.5%.

Along with the offering, we entered into capped calls, which increased the conversion premium to 100%. As such, until we reach an effective conversion stock price of $141.82, the 2030 note will not factor into non-GAAP dilution. And of course, no cash interest is due on the note. Net proceeds from the 2030 note, along with the issuance of 5.3 million shares and $3.5 million in balance sheet cash were used to fully retire our 2028 notes with a coupon of 4% and about $219 million of the 2027 notes with a coupon of 1.625%.

We also fully repaid our term loan for which we were incurring cash interest at about 5.8%. Our debt currently consists of $402.5 million of the 2030 notes and $100.5 million of the 2027 notes. Currently, annualized interest expense is under $3 million compared to $75 million for the third quarter of last year. Our significantly reduced cash burden from interest allows us to continue our acceleration of investments in strategic high-growth areas of our business while driving earnings growth and positive cash flow.

We were well positioned for the forward-sensing portfolio acquisition, which closed at the beginning of Q4 and which we expect will integrate very well into our sensing portfolio. Other net non-operating expenses were $0.4 million, primarily from foreign exchange revaluation losses. We recorded adjusted diluted earnings per share of $0.48, up 17% sequentially from $0.41 and up 85% from $0.26 recorded a year ago.

Our income statement results have translated to strengthening cash flow. Operating cash flow for Q3 was $47.5 million, sequentially up 7% from $44.4 million and up 60% from $29.6 million a year ago. Free cash flow for Q3 was $44.6 million, sequentially up 8% from $41.5 million and up 53% from $29.1 million a year ago. We ended Q3 with a cash and cash equivalents balance of $164.7 million. At the end of Q3, net debt sequentially decreased $20.8 million to $338.3 million.

Along with debt reduction, strong business performance contributed to an adjusted net leverage ratio of 1.5 as of the close of Q3, down sequentially from 1.6 and down year-over-year from 7.2.

Now, turning to our fourth quarter outlook. We currently expect net sales of $273 million, plus or minus $5 million, up 9% year-over-year at the midpoint. We expect net sales from our Infrastructure end market to increase sequentially, supported by projected sequential data center growth of approximately 10%. We expect net sales from our High-End Consumer end market to decrease about 3% sequentially. Typical seasonality in this end market is expected to be partially offset primarily by market share gains, but also from projected contributions from the newly acquired force-sensing portfolio.

We expect net sales from our Industrial end market to be about flat with lower decreases offset by growth in IoT systems and connectivity. Based on expected product mix and net sales levels, we expect adjusted gross margin to be 51.2%, plus or minus 50 basis points. Our consolidated adjusted gross margin outlook reflects sequential mix changes in the Industrial end market. We project strong net sales growth from cellular modules, which are part of our IoT systems and connectivity business with gross margins that are considerably below the corporate average.

Lower net sales with gross margins considerably above the corporate average are expected at about the midpoint of the $30 million to $40 million range we provided during last quarter's call. That said, our gross margin outlook for our total semiconductor products is expected to be 60.5%, plus or minus 50 basis points; up 220 basis points year-over-year at the midpoint and incorporates our projections of data center net sales growth.

Adjusted net operating expenses are expected to be $91.2 million, plus or minus $1 million, resulting in adjusted operating margin at the midpoint of 17.8%. Included in the higher fourth quarter adjusted operating expense outlook are R&D costs associated with the addition of the force-sensing business, along with increased investment in support of our growing data center portfolio. We have demonstrated strong returns on our R&D investment and expect incremental returns from our future investments.

Adjusted EBITDA is expected to be $56 million, plus or minus $3 million, resulting in adjusted EBITDA margin at the midpoint of 20.5%. We expect adjusted interest and other expense net to be approximately $0.5 million. We expect an adjusted normalized income tax rate of 15%, consistent with Q3. These amounts are expected to result in adjusted diluted earnings per share of $0.43, plus or minus $0.03 based on a weighted average share count of 95.7 million shares.

Thank you, Mark. We can now turn the call back over to the operator for the question-and-answer session.

[Operator Instructions] And our first question comes from the line of Rick Schafer with Oppenheimer & Company.

Congrats to you guys. My first question, I guess, is really on CopperEdge. It sounds like it's ramping with your lead CSP this quarter. And I'm just really curious sort of -- to me, that's breaking the ice. I mean does that -- how does that set you up with other CSPs next year? Basically, this validation from this lead customer speed deployments or wins with other CSPs? And is there any sense that you have today or maybe it's just too early to know, but any sense of how many ACC or CopperEdge customers you expect to ship to next year?

Rick, thank you very much for your questions. So, for the CopperEdge in ACC to supporting the leading hyperscalers, they have the product designed into 3 programs and then anticipate the ramp to start in the mid of 2026. But we supply our CopperEdge ICs to cable manufacturers, so they certainly need the product earlier than that. And we right now having all the things ready to go and based on the forecast they provided, so we need to make capacity available to support the very rapid ramp. And in our Q4, our revenue is just starting for that customer, but the substantial ramp is going to be throughout the fiscal year 2026.

As for -- you mentioned about, yes, this sort of icebreaking adoption. Definitely, we view this as a catalyst because of the benefits of the ACC over the competing technology is very obvious, especially in the power savings. And so, when we engage with the different hyperscalers, we sense that ACC is typically adopted in a platform design. So, certainly, now we see the broad-based awareness of this advantage and deployment by this hyperscaler certainly will provide a strong reference point for other hyperscalers to use in their future platform design. As for how many, we have been working with our cable partners and engaging pretty much with every hyperscalers out there. I do expect more design wins in the coming quarters.

And maybe for my follow-up, it's just a question I think we all get a lot and you do too, I'm sure, is just sort of how do you approach sizing the ACC opportunity? I mean, is a good proxy sort of the roughly $100 million DAC cable market? Or I guess, just sort of a starting point, some way to kind of size that market. And as part of your answer, I'd be curious just to understand better where -- we clearly see the benefits in terms of latency and power of ACC, but where -- like what kind of workloads or what kind of designs, where does ACC win versus AEC? Like where some of the lowest hanging fruit for Semtech there?

Yes. So, what we see, the ACC is positioned in a sweet spot between DAC, which had a signal limitation -- signal integrity limitation for transmitting over a longer distance at a high speed. Then AEC, which certainly can transmit with a longer distance, but with a significantly higher power consumption. So, certainly, if you look at the AEC plus DAC, it's a huge TAM. And ACC, as I said, is so uniquely positioned, we'll chip away a substantial portion of it as we start deploying hyperscalers. So, over time, we'll have a better idea how do we quantify the opportunity in the future.

And our next question comes from the line of Sean O'Loughlin with TD Cowen and Company.

Congrats on the solid results here. You mentioned -- Hong, you mentioned Q4 growth in data center. I think you used the word meaningful contribution from LPO in the quarter. Maybe you could just either talk about that deployment specifically or if you can't get into any details onto that deployment, but in general, how do you envision LPO coming to the market? Is it sort of like on the ACC side where it's very project specific and therefore, kind of concentrated and lumpy? Or do you sort of envision it to fold into the mix over time like a CPU server chip of all would have been into the new generations?

Sean, thank you for the question. So, we see a strong sequential growth opportunity in Q4 for our data center business. As we mentioned, we anticipate approximately 10% quarter-over-quarter growth. That's on top of 8% sequential growth from Q2 to Q3. And the majority of the growth is going to be on the FiberEdge product.

In LPO, we are gaining more hyperscaler design wins. We anticipate a meaningful contribution for LPO for Q4. If I have to say meaningful mid-single-digit level. And ACC contribution of Q4 is still going to be very nominal. As I mentioned early on, that the hyperscaler ramp in volume and their racks for the interconnect is going to be starting in the mid of 2026, will probably be 3 to 4 months prior to that, getting the ramp going to supply to cable manufacturers to get the ACCs ready.

Great. And then I just had a quick question for Mark. On the gross margin side in Q4, understood on the mix shift within IoT. I guess sort of 2 questions around that. Is this sort of a permanent mix shift that you're anticipating? Or is this just sort of a temporary -- LoRa was strong in Q3, going to dip a little bit in Q4, and so that will balance out longer term.

And then, well I guess maybe this part is -- I don't know if this is a [ soft ] question or a Mark question, but one of the suppliers on the foundry side in your space talked about some significant CapEx that they were anticipating on the silicon photonics, but especially the silicon germanium side. And just wondering if you anticipate any sort of headwinds on that side as you mentioned, the entire market goes through a bit of a capacity constrained environment here?

Thanks for the question, Sean. Let me try to address gross margin first and clarify there. On the positive side, semiconductor gross margins driven by data center and LoRa was 61.4% in Q3, up 140 basis points year-over-year, up 60 basis points quarter-over-quarter.

SIP gross margin, Signal Integrity Products, which encompasses data center, it was -- the gross margin for Q3 was 65.1%, up 270 basis points quarter-over-quarter -- 200 basis points year-over-year. And the reason I'm providing these statistics is we have -- what we've delineated as our core portfolio of data center, LoRa and PerSe, those are our faster-growing markets. And those gross margins, hopefully, you can see are above the corporate gross margin averages.

So as those businesses, we believe, will continue to ramp, we believe they'll be accretive to gross margins. On the IoT systems and connectivity side, we mentioned that we're going to have growth in cellular modules, where gross margin is less than 1/3 of our semiconductor products. And we also have nominally lower sales from LoRa.

In terms of where that's heading, we've talked about what's in our core portfolio and what is not in our core portfolio. I think from what Hong mentioned, one of the top priorities is, we're looking at portfolio rationalization there, partially to remove that margin disparity.

On your question on foundry...

Maybe I can answer that one. Yes, Sean, you are right. The silicon germanium technology platform right now is widely used in the making of physical media devices, the TIAs and drivers, but also in silicon photonics. That is why my first priority over the next few months is to make capacity available. That foundry is our close partners. We have long-standing relationship over the last 1.5 decades, and they have been providing excellent support to us.

Another nuance related to the component availability is a co-planning process with our customers and with our customers' customers sometimes. And that is a process we have implemented a few quarters ago and has been working extremely well. So even they share their visibility even in the business development stage. And we -- based on the understanding and based on our triangulation, we have the wafer start in the fab. And typically, the lead time is 6 months plus/minus. But with the planning process, we gained the visibility, we can start earlier, and we have never really let our customers down with our key components.

And our next question comes from the line of Harsh Kumar with Piper Sandler.

Hong, on last call, I think I was giving you a little bit of a hard time on lack of sequential growth in the data center business. And I think suffice to say, you fixed that going forward. But I did have a question on that. My question is, LPO seems to be a little bit earlier than ACC, and it seems to be coming on. You're excited about it. But almost everybody I know struggles with the scope and size of that market. So, maybe if I can ask you again or not again, but if I can ask you to just help us understand how big can LPO be as a business? And maybe what is your positioning in the LPO business? And then as a follow-up on the ACC side, I wanted to ask you, for ACC, are you seeing applications that replace ACC -- I'm sorry, AEC or just brand-new applications?

Yes. Thank you, Harsh. So, first of all, on the data center growth, yes. So, we're seeing very strong momentum and the booking and outlook and forecast is very strong throughout the 2026. And the LPO is -- we're pretty excited about the first meaningful ramp in our Q4. And as I mentioned before, the LPO give us an incremental opportunity to bring more content in the transceivers. So we benefit from a strong backdrop of the transceiver demand with the retimed solutions where we provide market-leading TIAs.

By offering LPO, we have opportunities to offer drivers in addition to TIAs. So that will increase our TAM by 150%. So, we certainly welcome that transition. And definitely, the rollout of LPO will cannibalize the DSP-based solutions, but we'll do it any day with the increased TAM. ACC, on the other hand, is a net gain. So, right now, with the air pocket we're experiencing from the early adoption in the rack interconnects, the ramp with the hyperscalers is going to be giving us an acceleration of data center revenue.

The adoption dynamics for LPO and ACC is a little different. LPO tend to be gradual because in their switch fabric, they can -- as soon as they -- as long as they have the confidence in the signal integrity on the host they plug into, they can use LPO. ACC, on the other hand, is more the platform-based when they design the new rack platform, and they will have the power consumption envelope and ACC provides 90% of power saving compared to AEC. That is a huge amount.

When you look at the rack design right now, any rack, they probably have anywhere from 100 to 200 cables inside. So, each connector -- each cable have 2 connectors on each end, it can translate into significant power savings. So, ACC is encroaching into the established AEC market, but also the [ ACE ] market. And because they're short, connects are predominantly DAC-based. But with the ACC availability, especially for 200 gigabit per lane and ACC is expected to be mainstream for longer than meter reach in the future.

Understood. Very helpful as always. And my next question was the force sensing acquisition. I don't know much about it. Maybe you could tell us just really quickly, given this is an earnings call, just really quickly, what the product does and how you intend to use it? And how much revenue and kind of OpEx you had because OpEx jumped up quite a bit.

Yes. So the force sensing is a capability you need to touch it and to activate it. We have the capacitive sensing that basically when you have your human body close to the sensors, you change the dielectric constant, you can activate the sensing. But with the force sensing, you need to apply the force to activate it. It combines with capacitive sensing very nicely to offer broader capabilities for smart wearables and computing and automotive platforms.

The asset we acquired from Qorvo was originally as a company called NextInput, they were founded in 2012 and about 4.5 years ago, it was acquired by Qorvo. And the technology is very differentiating. They have over 175 patents issued under applications. When we were looking at how to grow the core asset and how do we fill the gaps in capabilities, the force sensing was on our road map. And just optimistically, we found this asset available. So we got them acquired and get them nicely integrated.

The acquisition happened slightly less than a month ago. But as I mentioned, the integration already been very successful. So we made the first shipment of the product with our fulfillment infrastructure last week. As for the incremental R&D increase, it's still a lot better to buy this asset than otherwise internally investing. And so it's largely a technology tuck-in. The revenue contribution at this point is immaterial, but we do project a very healthy synergy and a very healthy contribution of this technology and asset to our future revenue.

And Harsh, just to double-click on OpEx. While there is incremental OpEx from our sensing portfolio, including this force sensing product, we're also increasing R&D in data center. So the areas where we're investing, the core areas and these core assets isn't changing. And we've been able to deliver some pretty good returns in data center, LoRa and sensing with some nominal increases in OpEx. We expect to be doing the same in Q4.

And our next question comes from the line of Christopher Rolland with Susquehanna International Group.

Congrats on the results. I guess, first, a clarification and then a question. The first clarification is you talked about a customer integrating linear equalizers on PCB in the coming quarters. If you could talk a little bit about that. Is that a high-volume win or more of a test case? And then just a clarification, you said that you're ramping LPO with several leading U.S. hyperscalers. Are those the same 2 that you were talking about last quarter? Or are there additional for LPO?

Yes. Thank you, Chris, for the questions. First, on the customers are evaluating the linear equalizer in the PCBs or connectors. Those are for the high-volume applications for the high-speed traces and some customers are planning to do it in the PCBs. And some other customers are seeing the marginal loss of signal integrity from the host, from the ASIC to the port. So they are evaluating to use the linear equalizer to bridge the signal integrity. And it's pretty exciting. So it is pretty broad-based and more than 3, 4 customers on those type of applications.

As for the LPOs, we stopped counting how many hyperscalers are planning to use it. I would say, at this point, the conversation almost same when I was talking to the teams and when I was engaging with our customers doing the optical modules, the CIOE in Shenzhen and ECOC in Copenhagen, and we were joking, it's more like who are not planning of using LPO and why? So I would say this technology at this point is not if, but more like when and what platform they're going to be using it. And we're really excited about the Q4 mark the starting point of the ramp, but we do expect acceleration throughout 2026.

That's fantastic, Hong. And then secondly, I'll leave it up to you at dealer's choice, either PON in China and when we should get confirmed tenders and what you're hearing there or LoRa kind of your outlook there? It seems like this Gen 4 has some new use cases, which is pretty cool. You can answer either or both.

LoRa, one question on LoRa. So yes, Gen 4 is gaining tremendous momentum and the multi-protocol is really very exciting. So we plan to provide SDK and software stack to enable Wi-SUN first and to enable a security application combined with LoRaWAN.

And our next question comes from the line of Tim Arcuri with UBS.

Hong, your tone on divestiture has definitely changed versus what it was 3 or so months ago. It was sort of put on hold a little bit. And now it sounds like you have another adviser, and you have some folks who are interested. And so can you just -- like what changed? And was it you weren't getting the price you wanted and now these buyers are more interested in engaging at a price that you're happy with? What -- can you just walk through like the evolution of what sort of changed there? And maybe how close are you to -- do you think executing something?

Thank you for the question, Tim. So the simple answer is nothing has changed. So this time around, we definitely have more dedicated mind share from the potential acquirers and because the geopolitical situation is a little bit more settled and also they're seeing some of the tailwinds playing into the reality into the new business opportunities, the backlog and also we projected Q4 sequential growth. So to the right acquirers, this really representing a pretty significant synergistic value to them. And as for the timing, we really cannot predict, but rest assured, this is my top priority.

Okay. And then just on the rack. So it sounds -- I mean, you're kind of -- it sounds like you're semi-promising Kyber in 2027. And I just want to make sure, do you have a lot of visibility on that, just given what happened this year with the Blackwell racks. I just want to talk through how much confidence that you actually have on that, that you would be ramping on Kyber in '27 because it does sound like you're kind of semi-promising that.

Tim, I would shy away from the specific platform, but go back to the fundamentals. So there's really not a whole a lot of different ways to improve the signal integrity, especially when you get the high-speed signal launched into very thin metal trees. So you're going to lose the signal strength, you're going to distort the signal, and you need to condition it, and there is no better or more seamless way than getting linear equalizer integrated on the board. If they have other ways to do it, they would do it as well. So that's the fundamental belief and we have -- and that's the use cases we're seeing with multiple customers who are interested in incorporating linear equalizer on PCBs.

And our next question comes from the line of Tore Svanberg with Stifel.

Hong, my first question is on ACC. You mentioned the 3 programs there with the lead customer ramping in '26. I know you can't talk about specifics, but could you at least confirm that all 3 programs are either cable or PCB board? And are they all based on the same speeds? And if so, what are the speeds?

Tore, thank you for the question. All 3 programs are for 200 gigabit per second trace, and they are all in the cable forms, these 3 programs, none of them are the chip on board.

Yes. And as my follow-up and sort of back to the Sierra Wireless gross margins. They've been under quite a bit of pressure. I mean I understand the mix of modules versus services and so on and so forth. But we're also hearing about component costs going up, whether it's memory or modem chips or anything like that. So how should we think about that gross margin, not just next quarter, but over the next few quarters?

Yes. So Mark, do you want to...

Yes. So on the gross margin for the ISC business, so you mentioned memory. Maybe I'll just get to the point in terms of, let's say, inflationary costs on the BOM, we're not experiencing that, especially in memory. We have a few choices there in terms of suppliers. So there -- for ISC, it really is mix. That's the primary driver of gross margins. As we -- as cellular modules is a higher percentage, the gross margin goes down. And if we have more in terms of services or router business, the gross margin in that business goes up. But at this point, as we're guiding next quarter, we're seeing really, really strong orders and expectations for customer delivery ramps into Q4.

So Mark, so this is basically 5G modules that are ramping this quarter and because they're lower margin than services and software, that's basically what's weighing on it.

That's right. But it's both 4G and 5G. But of course, 5G is definitely a tailwind, correct.

And our next question comes from the line of Quinn Bolton with Needham & Company.

I wanted to follow-up on that last question. Just, Mark, maybe you can level set us. I think you said the semiconductor gross margin for the fourth quarter would be 60.5%. I don't know if you gave an ISC gross margin, but it looks like it's got to be pretty materially below the 36.6% that you did in the third quarter. So just wondering if you could give us some range where you think that ISC gross margin comes out in the fourth quarter?

Yes. At this point, I'll just say that the ISC gross margin is -- will be lower, primarily due to the drivers we talked about with cellular modules. On the -- but again, on the positive side, semiconductor gross margins was 60.5%, plus or minus 50 basis points. So still quite healthy, and we expect that to continue to grow. Well, we're only guiding on 1 quarter, right? The drivers of gross margin between data center, LoRa and per se, our sensing business now is expected to be accretive to that gross margin.

Got it. And then I'm not sure if it's for Mark or Hong. It does sound like you may be getting closer to a potential divestiture based on your comments in the script. I think in the past, you described a divestiture of noncore assets as being non-dilutive to EPS because you would take deal proceeds and pay down high interest rate term loan debt. Well, you've now done that interest expense annually is less than $3 million. So I guess I'm hoping you could comment now without the balance sheet, would a divestiture of noncore assets be dilutive to EPS?

Yes. At this point, we're looking at the lower gross margin portions of our business. So that would be something that -- at this point, without naming all the specific assets that we're looking at, let's say it's a nominal impact to immaterial impact.

And our final question comes from the line of Cody Acree with The Benchmark Company.

Hong, if we can maybe go back to the ACC opportunity for a minute. Can you talk to some of the market's concerns around reliability of ACC and earlier testing? Is that contributing to any of the hyperscaler what looks like to be maybe incremental delays in the program ramp that I believe was expected to begin here in Q4 and now looks like it's more into next year.

Cody, yes, thanks for the question. I'm not aware of any reliability you were mentioning about the ACC. So there are some chattering like this is more like a couple of years ago by not really -- the player is no longer active in the industry, and there's some false start. But there are no -- any issue we are aware of, and we have deployed a significant amount of ACCs in the industry. So I mean, we haven't heard anything bad. And then with the hyperscalers, they have gone through months of qualification and reliability testing, system validation process. They're very careful and they're very technically capable. We haven't heard anything -- any concern about that.

All right. Great. Can we go back to your discussion of ensuring capacity availability? Can you just talk about some of the strategies that you might be able to employ specifically on the wafer side that you mentioned earlier?

Yes. So we talk about silicon photonics, silicon germanium semiconductor platform to support silicon photonics and also our PMD physical media devices. And there are more than 1 fab, one location, and we try to make more capacity available by engaging and qualifying manufacturing from other sites in other countries. So that not only unlock some additional capacities, but also make it more robust from the geopolitical point of view. So that's what I mean. That's the focus for the near term.

Would you look at anything like dedicated capacity commitment or investment on your part?

So the capacity, certainly, we have been increasing the CapEx investment in the back end, for example, testing. And -- but for the foundry capacity, we are primarily working with our partners to qualify their foundry from different locations.

Thank you. And with that, there are no further questions at this time. I would like to turn the call back over to Mitch Haws for closing remarks.

That concludes today's call. Thanks to all of you for joining us today. We look forward to seeing you at various investor events over the coming weeks.

Thank you, Mitch. And with that, this does conclude today's teleconference.