Advanced Energy Industries, Inc. Stock

Everyone. Welcome to the last day of Citi's Global TMT Conference. My name is Elizabeth Sun, and I'm in the Citi Semi and Semi Cap research team. Today, with me is Advanced Energy. Joining me on the stage, we are -- we have Steve Kelley, President and CEO; Paul Oldham, Executive VP and CFO. And in the audience, we have Edwin Mok, Senior VP of IR. And before we begin, I believe Paul has a few comments regarding Reg FD.

Great. Thanks, Elizabeth, and thanks, everyone, for joining us. Just a reminder today, any statements we make are subject to a number of risks. You can find those risks articulated in our recent SEC filings. Also, we reported our Q2 earnings on August 3 of this year. And today, we will not be providing any updates to guidance provided at that time.

All right. So Steve, let's start with some questions on your semis business. So after delivering a meaningful semi upside in Q2, you raised your second half guidance for semi to grow almost 50% year-over-year. I think earlier this year, there was a lot of concerns around you being very kind of conservative on guiding the semis growth. So what changed over the past -- over the course of like past 6 months? And what gives you the confidence to revise up the semis growth this year?

Yes. Thanks, Elizabeth. I think what we saw in the first half was an increase in our customer confidence over the next 2 years and also an increase in their forecast. And so we saw a number of new orders that came in that gave us a lot of confidence in the second half as well as in 2027. So one of the actions we took in response to those increased forecast was to accelerate the opening of our new factory in Thailand. So in Thailand, we're just opening a 500,000 square foot facility, and we'll be launching that facility with our largest customers, semiconductor as well as our largest customers in data center. So that's a very good sign for future demand.

And we'll get to the facility part a bit later. But staying in the semi business, if we look beyond '26, you just touched a bit on '27 looks good as well. For components guys like you, you typically outperform in the WFE -- early part of the WFE up cycle. So if we look beyond '26, looking into '27, what will drive this kind of momentum for you to outperform WFE.

Yes. So what we're being told by our customers is they have good visibility for the next 2 years. That's a lot of visibility. And they're also telling us they're constrained by available fab [ space ]. So as their customers are able to accelerate the build-out schedule for new fabs for advanced memory, advanced logic, that's an upside for our customers, upside for us.

And normally, your customer like typically build inventory at the early part of an up cycle. But you talked -- I think you talked about like you are largely shipping to customer demand. I was wondering why this time is different, why the customers are not really building inventories?

Yes. I think it's an interesting phenomenon. We think we're shipping to demand and not into inventory. And I also look at how we manage our supply base. And we learned a lot during COVID about managing our supply base. We divided the suppliers into 2 categories, the reliable ones and the not so reliable suppliers. And so if you're not reliable supplier, we'll typically put in place more inventory to give us some insurance against execution challenges. I think the same phenomenon happens with our customers.

I think Advanced Energy is perceived as a reliable supplier. We've got very strong factory infrastructure. We're successful in managing our supply chain. So they have to put in place less inventory as insurance for shortfalls from Advanced Energy. So I think we're in the good supplier category. And therefore, our suppliers -- our customers don't need to put in place a lot of inventory.

And what is your lead time?

The lead times?

Yes, within.

Yes. So in our semi business, we typically operate on a just-in-time model with our largest customers. So there's no lead time per se, but we know what they need and when they need it, and we're able to keep up with that.

Good to hear. And I want to touch a bit on the new products we've talked about for the past 2, 3 years, eVoS, eVerest, NavX. I think feedbacks continue to be good. Can you remind us where are we on the ramp? And are you expecting them to start contributing more on the growth or share gains next year?

Yes. We're very enthusiastic about the customer acceptance of our new technologies, namely eVoS, eVerest and NavX. And where those technologies are used is leading edge. So today, we're being designed into sub-2-nanometer applications, logic, for instance. And so -- we expect those products to become more meaningful from a revenue standpoint next year. We're already seeing shipments this year, but compared to the overall shipments we make, they're not that meaningful. But we're very encouraged at the adoption of these technologies leading-edge logic and leading-edge memory because they're necessary to maintain throughput and yield at these very advanced geometries.

I'm curious within that ramp next year, how much of that is driven by continuous slot wins? And how much of that is driven by new like sockets?

Yes. So as you see the ramp this year, that's almost all older technologies. So they're existing wins from many years ago where our customers are ramping the volume. I think as you move into '27 and '28, you'll see a more significant percentage of the buy coming from new products as those new processes ramp to volume.

Got it. And in a longer-term question and as semi process become more complex, you talk about like sub-2-nanometer processes, you need more etch and deposition steps have higher aspect ratios and you need more maybe like demanding process control. So overall, how do you think that does that translate into the power demand and specifically for AE's products?

Yes. Obviously, higher etch and [ dep ] intensity is great news for us. It's great news for our customers, too. And the reason why there's higher etch and depth intensity is there's more process steps. So the end customers, the fab operators are trying to build very complex structures, very fine geometries. So this requires a lot of steps. and that drives the etch and dep intensity. But in addition to that, the technical challenges of dealing with these structures require higher power from our products, sometimes multiple frequencies and also more advanced metrology. So all these capabilities are built into our boxes, and they drive higher prices.

And just final question on the semi side. I think you talked about you're starting to gain some traction on the testing side. So can you talk about like what do you provide there? Is -- how is it different from the front end -- the front-end guys? And then what's your -- what's the expectation of the ramp here?

Yes. It's interesting. Our heritage has been in plasma power, very, very strong technical challenge, helping our customers plasma lit inside plasma chambers. But over the past few years, we put more emphasis on what we call our system power business. This is basically the box between the wall and -- and the demands have become increasingly complex, particularly as we move into the AI testing challenges. So we've had significant success there getting our products designed into those testers, burn-in ovens, other applications where our customers need reliable system power.

And should we expect this demand go up with -- to grow with kind of this testing or it's more about share gains?

Yes. We haven't modeled exactly how it's going to grow. But what we know is we continue to expand capacity every quarter for these test applications. So it's a very good business for us, and it drives margins similar to what we have in the plasma power business.

And switching to data center side, apparently, you guys are a data center company now. So data center revenue more than doubled in '25, and you're talking about -- you're guiding more than 50% growth in '26. So for '26, I'm curious, is the growth more from continued build-out in your top customers in top customer -- one top customer in data center? Or is it also a broadening of your positions in the other hyperscalers?

Yes. So in '26, what we've seen is a ramp based on our hyperscaler customer only. So we talked about a second wave customer set, but we don't see meaningful revenue from those customers until '27 and '28. But what happened in '26 was a number of our wins from 2025, the design wins. are ramping to volume. These wins typically carry higher ASPs because they're driving higher power requirements. And so it's basically our ability to execute with a chosen subset of hyperscale customers that's driving our revenue growth.

And just to that higher ASPs, can you help us understand like directionally, how does that pricing increase generation or like year-by-year, like give us a range of this pricing upside?

Yes. So the cadence right now is every year, a new solution is required. That's loosely tied to new introductions from NVIDIA and other suppliers. And so the power requirements go up every year. But the space that we have to operate in doesn't change. And so basically, we have to deliver higher power same size box. And that's called higher power density. So the challenge is, how do you do that, increase the power density, you maintain high efficiency, very high reliability. So that's the technical challenge. And that's where Advanced Energy shines essentially. So that's why customers come back to us because we deliver a very reliable, efficient and high-power density solution.

Got it. And you have one customer that maybe more than half of your data center revenue. Just wondering, are you continuing to gain new contents, new sockets at that top customer?

Yes. Yes, we focus on a subset of customers, and we're gaining share at those customers. And each of those customers has a wide variety of applications where we're designed in. So even though we're highly customer concentrated within those customers, we have some diversification across different applications.

And you talked about a second wave of customers earlier. How different is that opportunity from your existing hyperscaler business in terms of product requirements, customer concentration and content per AE?

Yes. So when we deal with hyperscale customers, they require a high degree of customization. So it's very engineering-intensive effort. That's why we've basically constrained ourselves to a subset of hyperscale customers because we don't want to get overextended there. I think that's been a good strategy for the company. As we look at the second wave customers, these customers will typically take standard products from us with some minor modifications. So the engineering intensity is much less for our second wave customers. And so this allows us to engage with them without overstretching our engineering.

And does that give you a higher margin?

What we see is the margin is very similar to what we have with our hyperscale customers. But obviously, since the engineering intensity is less, I think the operating margin is probably a bit better with our second wave customers.

And how big is the opportunity for the second wave of customers? I think you talked about it could be a similar size to one big hyperscale customer. But when do you expect to reach that level?

Yes. Yes, -- we have said that in aggregate, we think the second wave customers could be as big as one of our hyperscale customers. We expect to reach that probably sometime in the '28, '29 time frame. Hard to gauge exactly. But we're very enthusiastic about the second wave customers as a way to diversify our customer risk.

Good to hear. And I do want to ask on 800 volt. There is a lot of questions around this, and I'm sure you are receiving a lot of questions on this as well. The first is, as power architecture moves from 800 volt, how do we -- how should we think about the change in Advanced Energy's dollar content, both in the per rack level and also per megawatt level?

Yes. So we're heavily engaged with customers on 800-volt solutions. And the challenge is very similar to the challenge we currently have with 48 or 50-volt solution. It's all about power density, efficiency, reliability. And so it's just -- it's the next step for us. Generally speaking, every time an architecture changes, it's good for Advanced Energy because we're a technology leader. We can bring technologies to bear to maximize the efficiency, power density without sacrificing reliability. And so that's our value proposition. The way we approached 800 volts is through a modular approach to our products.

So we have different modules we've designed, then we can combine these modules in different ways to meet the needs of different customers. And that's working pretty well for us actually. So it's a mix and match strategy because the expectations and standards are changing pretty dynamic. But at the end of the day, we don't 800-volt to be significant from a revenue standpoint until 2028 at the earliest. So we think 2027 is a design win year for us and '28 and '29, start to see revenue on the 800-volt product.

And any early thoughts on your content?

Yes. We look at the content, and we think the content goes up like it has in every other generation because the power is going up, the challenge is getting larger from a technology standpoint. And so we think we're in pretty good shape to grow overall content, maintain or improve our gross margins.

Content that means content per rack or.

Yes, it'd be content per rack where we participate. I think it's important in data center to remember that our strategy is not to grow market share necessarily. It's to operate those applications that will drive higher margins for the company.

And something that investors are trying to figure out is when AC/DC moves from the rack to [ sidecar ] or power rack, people are trying to figure out if you guys have a solution for that and you guys are expecting -- or already engaging with customers on that and expecting.

Yes. So we do have solutions, and we're engaging with customers on the sidecar as well as more conventional architectures. So I think that's one of the advantages of our modular strategy. It gives us a lot of flexibility on how we deal with changing customer requirements. So I think we're in pretty good shape no matter which architecture a customer chooses.

And where are -- what will be the biggest content opportunities for you guys in this kind of transition? What needs to happen in order for AE to grow your content, like as you said, grow your content meaningfully in this one?

Yes. I think the 2 key parts of this market are, one, during the design phase. So it's having the right engineering team in place, having the right IP blocks in place, so you can move quickly. And the second is your ability to scale manufacturing quickly and reliably. And I think we've got a really good engineering team, and we need to add to that team, but it's difficult to find really good engineers. So that's the constraint. It's not our willingness to spend money on more engineers, just our ability to find them, but we are expanding.

As far as new factory space, I think we've done an excellent job in the Philippines and in Mexico, expanding capacity to meet the hyperscaler requirements. We've also done a great job upscaling our development sites because it costs more and more to develop these products with the higher power requirements. But looking to the future, we have this large factory in Thailand, 500,000 square foot factory, and that will be the future for next wave of expansion in data center.

And just to double-click into the -- talk about the kind of bottleneck will be the engineering innovations. So what are the most difficult things to do there? Is that efficiency, power density, like thermal management, reliability? Or is it really all of them?

It's all of the above. So that's the key. How can you do this? How can you juggle these different requirements and come out with a solution that's manufacturable, reliable and scalable. And that's what we've been doing for the past few years. So we've built a fair amount of confidence with our customers that we can get the job done. And one of the things we try to do from generation to generation is reuse as much as possible. So we're not inventing something new every year or at least we limit the invention, right? So I think that's the key is focus on a subset of customers and execute 100% on their needs.

Got it. And in the longer term, also -- there is also like the 800-volt directly to step down to 12-volt rather than using an intermediate 48 and 50-volt bus. So does that increase or decrease AE's content opportunity? And on the one side, on handset is just one big step down. On the other hand side is like it's a big step down. So like...

Yes, yes. And I think it depends on which customer you talk to what they want to do. And that also drove our decision to go to a modular approach with 800 volts. So the way to think about this is we have what we call a PDU, a power distribution unit. Think of it as a motherboard essentially, where we could plug in these different 800-volt modules. And you can do whatever you want with it, 800 volts to 6 volts or to some other intermediate voltage. So it's up to the customer how they want to deploy this PDU. And so that flexibility is really important at this stage of the game because I don't think anybody knows exactly how the architecture are going to look a year or 2 from now. But if you have flexible modules, you're able to experiment and figure out what makes most sense for you.

That's your strength.

Yes. And so we're all about customization.

I'll take a pause here to see if any questions. Okay. Let's move on to Industrial and Medical. So I think it finally looks on the way to some recovery and some growth there. How do you see the rest of the year shake up and then look forward to next year?

Yes. The industrial and medical market has recovered. There's significant increases going on. I think we were up 11% quarter-on-quarter, 16% year-on-year in Q2. And as we look forward, we see each quarter getting stronger in '27. And we're seeing anything AI related, let's say, physical AI is driving a lot of the high-end industrial medical business. And we operate in the high end of that business. So it's factory automation, robotics, test apparatus, those types of applications. And that part of the market is definitely strong. And we're basically reaping the benefits of our design wins over the past 2 to 3 years in industrial and medical. A lot of those are starting to ramp the volume. And so we're basically expanding capacity rapidly in the Philippines and in Mexico, where most of these products are built.

And I think it was in Q1, you had to make for lack of better words, sacrifications to the industrial medical because other products are ramping a lot faster. Our demand is coming like got pulling a lot faster. So at this stage right now, do you have enough capacity to support like the ramp across your segments?

Yes, yes. So we had a hiccup in Q1. It was really our own problem. We had an execution problem in the company, which we've rectified. But we have enough capacity in industrial medical. We've basically corrected for execution problems. So I think we'll be caught up with our delinquent backlog by the end of this year in Industrial and Medical. But there are no limits right now as far as what we can invest in industrial medical capacity.

And I feel like this like telecom and networking was never really asked a lot. So how strategic is this T&N segment market to your long-term target?

We're starting to see more opportunities that are AI-related telecom and networking. So I'd say that definitely works into our strategy. But as a market, it's -- I wouldn't say it strategic. It's more opportunistic for us. So where we can reuse our existing technology to satisfy the needs of telecom and networking customers, we will. But we're not going to develop ground-up products for that segment.

And what kind of AI opportunities or maybe just overall, like what do you provide there?

It's basically the same -- very similar products to what we provide to the data center customers. And so you see a convergence there with different companies that need the same basic thing, which is reliable power, highly efficient power, high power density operating.

And some questions for Paul, maybe. On the capacity side, maybe can you explain your capacity expansion strategies? You talk about like [ Thailand ] or expanding a lot of capacity right now. And your customers are also aggressively adding products capacity as well, and we heard a lot of doubles over the past couple of days during our conference. So will you be able to keep up with demand if they grow more than -- the demand grows more than expected? And on the other hand, do you have any concerns for putting too much investment in the capacity?

Yes, it's a good question. We've clearly biased ourselves to make sure we have enough capacity, both in terms of physical capacity and in materials to capture upside. I think we've said that for a while. We do see a ramp going on both data center and the semiconductor market, and we're committed to make sure we can capture those opportunities. That means we've carried more inventory, and we're investing quite a lot more in capital expenditures. Good news is we started this a while ago. We saw this trend coming, and we're able to get in front of it.

As Steve mentioned, we started our Thailand facility build-out back in 2023. So that factory is now coming on stream right when we need it with an ability to scale it very quickly. Today, we have over $3 billion of capacity in our existing facilities with the opportunity to potentially expand that more adjacent space through how we configure things, what we put in those manufacturing line. But with Thailand coming on, that would give us over $5 billion of capacity as soon as 2028. So we feel like we're in good shape from that perspective. In terms of what's the risk of like overinvesting, I think there's a few things to think about.

First of all, the payback on this investment is very fast. Secondly, a lot of these investments are very fungible because we're in multiple markets, and we're basically doing final assembly and test, we can move this equipment around depending on what's the actual need at the time. And the third thing is our investments, while they're significant for us, we're not building a fab or a back-end testing facility or something like that. And so the relative capital investment is fairly small. So all in, we believe that these investments are already built into our business model. In fact, if anything, all we've done is we pulled forward investments that were already in our 2030 model because, frankly, the demand is running ahead of what we expected.

Got it. And with that $5 billion capacity with [ Thailand ] ramp up, I think maybe a quarter ago or 4 months ago, it was -- the number was kind of $3.5 billion to $4 billion. So what drove that upside? Where does it come from?

I think it comes from a number of things. I think, first of all, we sharpened our pencil a bit. I think capacity is never one number. It's sort of a range. I think you're trying to look at that more on that context. But part of it is accelerating investments. We've increased our capital expenditures to capture -- make sure we've got equipment and space needed to meet the growth ramp.

And while [ Thailand ] ramping, -- is there a period should we think about there kind of underutilization cost? And ultimately, I believe the Thailand facility is margin accretive.

Yes, it's a good question. The investment in Thailand and the other investments in capacity were always in our 2030 business model. So it's not new or it's not incremental. What's happened is we pulled that in because we're much further ahead. We're almost getting to the revenue levels in 2027 that we had in our model we put out just years ago, which was a 2030 target. So we pulled that in. So it's contemplated in the model. And in fact, this last quarter, we said we expect gross margins to exit this current year around 42% and next year around 43%. That also contemplates the investment in bringing Thailand and the other manufacturing investments on stream.

And just on the gross margin, talk about you are very close to your target model right now, but gross margin is just a bit below. I think there is tariffs impact in there. There is like ramping costs you talked about. But is there anything else that we should think about that is kind of impacting the gross margin, not reaching the target?

You hit the big ones, right? Obviously, tariffs were not embedded in our original model. So that's a headwind. Also, I think the mix of being more data center heavy is a little bit of a headwind. But frankly, margins are improving across all of our markets. And so we wouldn't be able to continue to improve margins if data center margins weren't also improved. And so when we look forward, I think the important thing is, yes, we're approaching our model from a revenue perspective. We're also approaching it from a gross margin perspective. And we believe that the underlying factors that are improving gross margin can continue.

We still expect the majority of the benefit from mix of new products is ahead of us. So that's a positive. When you ramp this quickly, it's not always as efficient as you'd like it to be to your last question. So there's still opportunity to wring out some efficiency in manufacturing. And we should continue to reap the benefits of volume. So while we put out a model that said 43.5%, we think we can certainly go beyond that as the business continues to grow and our new products actually become a bigger proportion of the total revenue.

And just to that point, your Analyst Day was like '24?

Yes, just 2 years ago, not even 2 years ago.

And are you expecting to update your target model anytime soon?

Yes, -- it's a good question. Ultimately, we will do an update to it. But I think the beauty of the model is it's -- when we put in place, it's a scalable model, which says that there's an opportunity to actually continue beyond kind of what would be the bookings of that model. Clearly, revenue is tracking ahead, data center has been much stronger than expected. Semi is in a fast ramp and I&M is finally recovering. So I think those things bode well for continued growth. We think the elements of gross margin outside of tariffs and a little bit of mix are structurally intact, which can also go beyond the 43.5%. And frankly, we're quite a bit ahead on operating leverage because we've grown faster. So we think while we haven't updated the model yet, the underlying tenets of the model still hold.

And just you are doing so well in semis and data center and industrial and medical is recovering. But when you think about investment across the segments, how do you balance this kind of making investment in semis and data center or like doing like some M&As in the industrial and medical?

Yes. So the way we think about it is we've really doubled down our investments on semiconductor and data center. Those are largely internal investments. We're invested significantly in inventory to put ourselves in a position to capture upside and be ahead of the ramp. We're investing in CapEx. We're spending roughly 2x the CapEx we would normally spend. And so while that's running high today as we get through these ramps, that can move back to a more normalized level of 3% to 4%, but we're certainly making those investments today. We think from a capital allocation perspective, M&A certainly is an important part of our strategy.

We think that, that's most tuned to the industrial and medical market, where it's a more much broader-based set of customers and set of competitors. And it's a way to use -- M&A is a way to build a bigger beachhead in that market because ultimately, we'd like to have 3 very strong markets that we address going forward.

And just to wrap up, is there anything that you think investors are not understanding Advanced Energy story very well or anything that we did important that we didn't really touch help you?

Yes, I think the key thing is that we have multiple legs of growth. If you look at our customers and the market dynamics, that growth is projected to continue. It may not always be linear every quarter, but it's projected to continue over the next couple of years. And we're tied right into those leading-edge technologies that support it.

No, thank you very much.

All right. We are out of time. So thank you very much for coming, and thanks, Steve and Paul for joining us today.

Thanks Elizabeth. Thanks, everyone.

Greetings, and welcome to the Advanced Energy Second Quarter 2026 Earnings Call.

[Operator Instructions]

As a reminder, this conference is being recorded. It is now my pleasure to introduce Edwin Mok, Senior Vice President of Strategic Markets and Investor Relations. Please go ahead.

Thank you, operator. Good afternoon, everyone. Welcome to Advanced Energy's Second Quarter 2026 Earnings Conference Call. With me today are Stephen Kelley, our President and CEO; and Paul Oldham, our Executive Vice President and CFO. You can find today's earnings press release and presentation on our website at ir.advancedenergy.com. Before we begin, let me remind you that today's call contains forward-looking statements. They are subject to risks and uncertainties that could cause actual results to differ materially and are not guarantees of future performance. Information concerning these risks can be found in our SEC filings. All forward-looking statements are based on management's estimates as of today, August 3, 2026, and the company assumes no obligation to update them. Any targets beyond current year presented today should not be interpreted as guidance. On today's call, our financial results are presented on a non-GAAP financial basis unless otherwise specified. Detailed reconciliation between our GAAP and non-GAAP results [indiscernible] press release. With that, let me pass the call to our President and CEO, Steve Kelley.

Thanks, Edwin. Good afternoon, everyone, and thanks for joining the call. In the second quarter, we delivered record results with revenue and earnings above the high end of our guidance. Demand strengthened in all of our target markets and solid factory execution allowed us to capture upside within the quarter. Looking forward, we expect record revenue in both Q3 and Q4. Our investments in capacity and piece part inventory are allowing us to keep pace with increasing customer demand. We continue to make progress improving gross margin.

In addition to manufacturing efficiency improvements, we are benefiting from a richer product mix. The success of our new products which deliver more value to the customer is the key reason our mix is improving. In addition, we have taken pricing actions in our mature product lines, primarily to offset the impact of increased input costs. We are increasing output in our factories in Malaysia. We are working closely with some of our largest data center and semiconductor customers to qualify our new Thailand factory where we now expect first production revenue in the fourth quarter.

When Thailand is fully built out, we expect to have roughly $5 billion of revenue-generating capacity across our factory network. On the new product front, we continue to increase our investment in R&D, which is critical to maintaining our technology leadership and competitive edge. Our modular approach is driving technology reuse across the company, reducing development cycle time and speeding time to market.

Now let me provide some color on each of our markets. In semiconductor, we delivered record revenue in the second quarter. The semiconductor equipment market is growing to record levels, driven by strong demand for leading-edge memory and logic as well as increasing etch and depth intensity. We are executing well to meet this increased demand, delivering sequential revenue growth of over 27% in the second quarter.

In Plasma Power, we continue to work closely with customers to tailor the performance of our eVerest and eVoS platforms to the requirements of next-generation processes. Customers have validated meaningful yield and throughput performance at the leading edge, and we believe that adoption of these platforms will drive market share gains for AE in the coming years. In system power, we secured design wins in test, ALD and thermal sensing applications in the second quarter. Across both plasma and system power, we have several programs beginning to ramp to production.

With additional design wins in the pipeline, we expect new product revenue to accelerate our revenue growth in 2027 and beyond. In data center computing, we see robust demand in the second half and now expect full year revenue growth of at least 50%. Our improved outlook is due to increased demand from hyperscalers. We continue to pursue multiple opportunities with second wave data center customers defined as customers outside of the top hyperscalers. We expect that these second wave customers will accelerate our revenue growth in 2027 and beyond. We have developed multiple products to address the power requirements of 800-volt data centers.

Our low-profile, modular solutions feature industry-leading power density, efficiency and reliability. By leveraging common modules, we are able to quickly develop customized products as this technology evolves. We have received positive feedback from a number of customers who have evaluated early production units of our 800-volt products. And based on customer road maps, we expect these products to go into high-volume production in 2028. In Industrial & Medical, revenue increased sequentially and year-on-year. Key design wins are beginning to ramp to volume and we see much healthier demand in the overall I&M market.

In the distribution channel, which accounts for roughly half of our I&M revenue, resales, orders and inventory all improved this quarter. On the execution front, we are working down our overdue I&M backlog and expect to catch up the demand in the second half. In medical, we secured multiple wins in therapeutic, imaging and life science applications. Our momentum in electrosurgery is particularly strong, fueled by our highly differentiated pulse power technology.

In industrial, we won key slots in test and measurement, factory automation and robotics applications. We also recorded multiple wins in aerospace and defense, with recognized versions of our leading-edge commercial products. We continue to leverage our digital marketing and channel strategies to broaden our I&M customer base. In the first half of 2026, design wins, which started as website inquiries increased 40% year-on-year. In telecom and networking, several customers are evaluating our rack power solutions for AI-related applications. We classify these as second wave data center opportunities, where we could redeploy existing technology blocks into new applications.

Now I'd like to update our view on 2026. Demand continues to strengthen, and we are executing to capture near-term revenue upside. As a result, we have increased our 2026 growth outlook to the low to mid-30% range. In semiconductor, we expect second half revenue to grow nearly 50% year-on-year. With our new products just beginning to ramp into production and additional wins in the pipeline, we believe that AE is well positioned to outgrow our market and gain share in the coming years. In data center, we are on track to increase revenue more than 50% in 2026 after more than doubling in 2025. We continue to win new programs at our hyperscale customers and are engaged with multiple second wave customers, setting us up for a strong 2027. In Industrial & Medical, we expect revenue to grow sequentially over the next few quarters. Market conditions have improved in a number of new product wins are ramping to volume.

Now for some closing thoughts. First, we are well positioned to deliver strong revenue growth into 2027 and beyond. Our design win pipeline is impressive, and our new products continue to hit the mark. We are making the necessary investments in capacity and inventory to allow us to keep pace with strong customer demand across all of our markets. We are fortunate to be a leading player in 2 large and fast-growing markets, AI data center and semiconductor.

Second, we continue to improve gross margin and have line of sight to over 43%. Finally, we continue to actively pursue potential acquisitions that make strategic and financial sense.

Paul will now provide more detailed financial information.

Thank you, Steve, and good afternoon, everyone.

We delivered record financial results in the second quarter. Revenue of $574 million increased 30% year-over-year, and EPS of $2.74 was up 83%, both exceeding the high end of our guidance. Gross margin improved 380 basis points year-on-year to 41.9%, and we delivered record operating income. Operating margin was well over 20%, a level that we've not achieved in many years. Operating cash flow increased meaningfully to $86 million.

Now let's review our second quarter financial results in more detail. Total revenue of $574 million was up 12% sequentially and 30% year-over-year. Solid execution allowed us to capture upside to our guidance. Semiconductor revenue was a record $278 million, up 27% quarter-over-quarter and 33% year-over-year. We believe that we are shipping in line with underlying demand. Data center computing revenue was down 1% sequentially as expected, but increased 35% over last year to $192 million. Demand progressively improved in the quarter as customers resolve downstream constraints setting up a higher second half. Industrial & Medical market revenue was $80 million, up 11% from last quarter and 17% from last year, as we increased output to meet customer demand. Telecom and networking revenue decreased 4% sequentially but grew 12% year-over-year to $24 million.

Gross margin in the second quarter was 41.9%. During the quarter, we received the large majority of our expected IEEPA tariff refunds. Excluding the benefit of these refunds, gross margin was above our guidance range at 40.7%, driven by higher volume and favorable mix of new products. Relative to Q2 earnings, the refunds benefited EPS by $0.04. Operating expenses of $115 million were slightly above our guidance range due to timing of program spending and variable costs. OpEx increased 11% year-over-year well less than half the rate of revenue growth.

As a result, second quarter operating income was a record $125 million, and operating margin increased 730 basis points year-over-year to 21.9%. Depreciation was $11.3 million, and our adjusted EBITDA was $137 million, up 84% year-over-year. Other income was roughly $5 million versus breakeven in Q1, mainly due to higher interest income and lower interest expense on the net proceeds of our convertible note offering in May. For Q2, our GAAP tax rate was 18.2%, driven largely by the nondeductibility of certain costs to partially retire our 2028 convertible note and related bond hedge and warrant. Our non-GAAP tax rate was 14% below our target of 16% to 17% due to favorable mix of earnings and certain discrete items.

Second quarter earnings were $2.74 per share, up meaningfully from $2.09 per share in the previous quarter and $1.50 per share a year ago.

Turning now to the balance sheet. In May, we completed a $1.15 billion offering of 0% coupon convertible notes due in 2031 and redeemed $438 million of our 2.5% convertible notes due in 2028. As a result, total cash and equivalents increased to $1.4 billion. Net cash was $132 million at the end of Q2. We also announced our plan to redeem the remaining $136 million of the 2028 convertible notes in September. In Q2, net working capital was up 2 days to 123 days, driven by increased inventory offset by improved DSO and DPO. We are investing in additional piece part inventory to support our customers' ramp plans enable manufacturing flexibility and capture upside opportunities. As a result, inventory increased 10 days to 145 days with turns at about 2.5x. Days sales outstanding decreased 3 days to 63 days, and days payable outstanding increased 5 days to 85 days.

Despite increased net working capital, Cash flow from continuing operations was a record $86 million. During the second quarter, we invested $50 million in CapEx to expand capacity and capability across our factory network including accelerated production investments in Thailand. Finally, we paid $4.1 million in quarterly dividends.

Turning now to our guidance. With strengthening demand across all of our markets, Initial production ramp of several new product programs and solid execution, we expect to deliver record financial results in both the third and fourth quarters. We are forecasting our third quarter revenue to be approximately $640 million, plus or minus $20 million. We expect Q3 gross margin to be in the 41% to 41.5% range up from Q2, excluding the onetime benefit of tariff refunds. We expect Q3 operating expenses to increase to $120 million to $124 million due primarily to investments in new products. We expect other income to be approximately $5 million on improved interest income.

We expect our tax rate to be in the 16% range for the next several quarters. As a result, we expect Q3 non-GAAP earnings per share to be $3, plus or minus $0.25 on 41 million shares outstanding. For the full year 2026, we are raising our revenue growth target to the low to mid-30% range, up from the low to mid-20% range. In semiconductor, we expect second half revenue to be up almost 50% year-on-year. In data center, we are raising our full year revenue growth outlook from the mid-30% range to at least 50% on accelerated hyperscale investments and ramp-up next-generation programs.

Industrial & Medical, we expect revenue will continue to grow sequentially on higher demand and better factory output. We expect gross margin to improve in the second half with Q4 in the 42% range. We expect full year OpEx to be in the $470 million range. As a result, earnings per share is expected to grow meaningfully faster than revenue for the year. Finally, we now project our 2026 CapEx will be in the $180 million to $195 million range, up slightly from our previous outlook based on accelerated investments to support growth ahead. Despite higher capital spending and investments in strategic inventory, we continue to target 2026 free cash flow to be at or above 2025.

Let me finish with some concluding comments. Advanced Energy is uniquely positioned to capture the strong market trends in both semiconductor and data center. We are investing aggressively in technology development, capacity and inventory to support growth. We believe our wide range of best-in-class power technologies, global engineering teams, production capacity and scale will enable us to grow share and capture upside. At the same time, with our strong balance sheet, we will continue to pursue inorganic growth opportunities to further diversify our scope. Finally, we expect to continue to improve our financial model by expanding gross margin, driving operating leverage, growing earnings and delivering higher cash flow. With that, we will now take your questions. Operator?

[Operator Instructions]

Our first question is from Matthew Prisco with Cantor Fitzgerald.

I just want to start on the data center side. Can you help us some more color on the increased positivity you're seeing versus 3 months ago? Maybe how has customer conversations developed? How are you thinking about the mix today? And then you commented that you see accelerating growth into 2027. What exactly does that mean?

Yes. Thanks for the question, Matthew. Yes. What's changed in the past 3 months is that our customers have been successful in removing some of the supply chain issues that we faced 3 months ago. So for us, that means we've been able to grow more bullish on our second half revenue and our 2027 revenue in data center. So this year, we've got a good line of sight to increasing business in data center. As we look into '27, we see a couple of key vectors there. The hyperscale customers who we've been engaged with for the last 3 years, we see a lot more wins ramping to volume there. We also are bringing on the second wave customers. these are the wins that we've earned this year that we'll ramp to volume next year.

And then as a third vector, we think we'll see some business at interval towards the end of next year, and we see that ramping to production in 2028.

And then the second question on the semiconductor side of the business, obviously, trends improving here as well. Just would love to hear your thoughts outside of the 25% of your business linked to service and online implants. How are you thinking about the ability for AE to outgrow WFE in '26 and maybe in 2027? And how do we think about that kind of new product traction that you've been highlighting on the call.

Yes. I think right now, I think we're better positioned to gain share than at any other time in our history. And I say that because of the acceptance of our new technologies, both in plasma power as well as system power. So today, I think we're well ahead of the share gain targets we published at our 2024 Analyst Day about 18 months ago. So we're very happy with the progress. We think we have laid the groundwork to gain share in conductor etch, in dielectric etch and deposition. And we're working at the leading edge nodes across all those categories of logic NAND and DRAM. The reason why we're preferred is that we're offering yield and throughput advantages to the customers as they move to these very challenging geometries.

So today, we see strong pull from all of our major customers. We're winning wherever we're engaged, and our modularity allows us to spin off derivatives relatively quickly. On the system power side, we've gained some major wins in process tools and in tester applications. And system power differs from plasma power products in the development cycle time is much less. So we're basically gaining share in 2 fronts in plasma power and system power in the semiconductor area.

Our next question is from Joe Quatrochi with Wells Fargo.

I was wondering if you could talk about your inventory situation and just you talked about buying more piece part inventory. How do you think about that in the forward demand? And I think going back, if you go back to the last kind of cycle, especially in semis component availability was a bit of an issue for you guys. Any sense of kind of where we sit today.

Yes. I guess the first thing I'll say is we learned a lot about our supply chain during COVID. So our team is battle-tested. And so what we did during the COVID shortages, we brought on second sources wherever possible. We build inventory where it was necessary. And for new designs, we stopped designing components from underperforming suppliers. Our current situation is we don't want to constrain our customers, and we don't want to constrain our revenue due to shortages of piece parts. So we're leaning into strategic inventory as lead times extend for these components and the demand that we see continues to increase. We think the inventory we're building right now is healthy inventory with very little obsolesces risk. We're also maintaining high factory staffing at our factories in the Philippines, in Malaysia and in Mexico. And that enables us to flex up when scarce parts arrive late in the quarter. So these actions that we're taking, both in inventory, the factory staffing are helping us stay ahead of our customer demand.

Man, I'll just add to that, Joe. We signaled going into this ramp a quarter or so ago that our intention was to lean into inventory because we fundamentally believe it would enable volume upsides and flexibility in manufacturing. And that's what we're doing. A couple of comments on that. We said that would impact turns in the near term, and we've seen turns come down to about 2.5, which is what we signaled. But I think it's worth noting that at the same time, we've invested in inventory to give ourselves flexibility and stay ahead of the curve, both AP and AR have improved, which mitigated the impact on working capital. So look, we have a strong balance sheet. We think it makes sense at this point in the cycle to invest in our own piece part inventory. As Steve said, to stay in front of customer needs and be able to deliver upsides.

And then maybe as a follow-up, as you think about just the puts and takes in gross margin. You talked about some price actions that you're taking. Any help on just quantifying how we think about that as part of the gross margin guidance?

Yes. First of all, I'd like to say we're reluctant to raise pricing unless it's absolutely necessary. So we first try to mitigate the price increases from our suppliers through second sourcing and through negotiation. But while we can't mitigate, we pass that through to the customer to preserve our own gross margin. Longer term, our first priority is delivering more value through new products, which tends to drive higher pricing and better margins.

Yes. Maybe I'll just add on that one, too, a little bit. If you think about that, our stated goal is to remain price cost neutral. I think that's working. Again, I think compared to a few years ago during the supply chain crisis, we're a little more in front of that and being proactive. If you look at what really drives gross margin going forward, it's a few things. Number one, in terms of future opportunity is, in fact, better mix coming from new products, as Steve mentioned.

The good news is we're already seeing some of that benefit come in. And there's more opportunity to run there. That's higher priced products, which have higher value to our customers and are tied to new products. Second, obviously, volume makes a difference. We've seen some of that this last quarter. We'll continue to see that as revenues grow. And finally, we've been growing at quite a rapid clip. And so we've certainly seen some inefficiency and manufacturing as we grow. We have high confidence that over time we'll be able to bring those costs out as well. So pricing, certainly in the near-term intention is to keep neutral, but we believe we can continue to improve gross margins really on structural improvements in the business.

Our next question from Krish Sankar with TD Cowen.

Two of them, first and almost congrats on the great results. I'm just wondering, when I look at your semi revenue guide, it's about a little over 30% year-over-year. Some of your customers are guiding to almost 40% revenue growth. I'm just kind of curious, has there been better inventory management from your customers that you're not seeing a sharper growth? Or is this the new normal where you kind of kind of grow more in line or slightly better than WFE, not outperform WFE when the cycle turns.

Chris, thanks for the question. So the first thing I would say is, over time, over the last 3 to 5 years, we have outgrown the competition. And so you do see variations year-to-year based on changes in the market where one particular product may be up 1 year or another product the next year. So it's hard to follow WFE every single year. But over time, we've done better than WFE and better than our competitors. But as I look forward, we see an extremely strong second half in semiconductor. So we think if you look at the second half of this year compared to the second half of last year, will be up almost 50% year-on-year. So no matter how you slice it, that's pretty good performance.

Got it. Got it. And then a follow-up on the data center side. You said that you're winning -- continue to win at hyperscalers. I'm just wondering how many hyperscaler customers do you have today on the data center power supply side? And along the path, the 800-volt DC. It seems like it's still like later next year into 2028 for your early revenue. Why is it late and some of your competitors who are shipping to AC hyperscalers or expecting early revenues in the second half of this year?

Yes, Chris, what we've said is that we have 4 hyperscale customers, but some of those are bigger than others. And they tend to be very engineering-intensive and we've been very successful with the hyperscalers where we [Audio Gap].

It's been a high level of interest in our solution at multiple customers. And they're low profile, and they've received very positive feedback from the customers who have sampled these products. The other feature is they're all focused on high efficiency [Audio Gap] in the 98% range. High power 800-volts going to coexist with both

[Audio Gap]

Is that a revenue contribution comment? And just what is the time frame in your vision for those second wave customers. [indiscernible] hyperscaler customers.

Yes, Shane, thank you for the question. [Audio Gap] to ramp the second wave customers. [Audio Gap]

Yes. I think it's difficult, if you look at our business to really find a metric where you could try to gauge our future revenue growth. I think what we've been doing is focusing on opportunities where we could be the sole source [Audio Gap]

And are those starting to [Audio Gap] extend to the broader architecture.

[Audio Gap] is that we've partnered closely with our key hyperscaler customers because there's very little time to perfect the solution. So what we've done is maximize reuse and work very closely with the customer to develop a solution that meets their next-generation needs. And so generally speaking, there's about a 9-, 12-month lag between the final design and the full ramp. And so this is a continuous cycle. Over time, what's changed is the number of projects for customers gone up as they develop confidence in Advanced Energy, both in our development capability as well as our ability to ramp the high volume.

When you say projects per customer, do you mean inside the 4 walls of an existing facility? Or does that extend to -- they're giving you more opportunities in other facilities as they grow?

Yes. When I said that, I really meant the number of projects that we have with a particular customer. And in many cases, we've qualified more than 1 facility to build these power supplies. And so really, we've got 3 choices now. One is the Philippines, one is in Mexico, and now we're bringing up Thailand. And so we'll basically build these products where it makes sense for the customer.

Got it. And then we talked a lot about increasing power density in the data center. But as you look at the leading-edge road maps in the semiconductor business, are you seeing power requirements per tool increase faster than unit volumes there as well?

Yes. So I think your question is, are we seeing increasing stress on power density and semiconductor fabs in addition to what we see in data centers. Is that correct?

Yes. Exactly. Like are you getting more content? Or is it a harder engineering problem at the leading edge as you look at the road maps going forward?

Yes. Yes, that's a good question. It's definitely a harder problem for a couple of reasons. One is the voltages tend to go up over time. And we see this across the portfolio as we move into the leading-edge development. The second is we don't have more room to get the job done. So typically, the space availability and the wafer fab is very constrained. And so our customers want us to accomplish the task in the smallest possible volume. And so we face significant challenges on power density. But we're able to basically use common technologies across our company. And so we apply learnings that we have made in data center to our semiconductor product development and vice versa. So I think this is one of the strength of Advanced Energy is we can apply learnings across all of our markets.

Right. So volume and content opportunities in both data center and semiconductor?

Yes.

Our next question is from Mehdi Hosseini with SIG.

Yes. A couple of follow-ups. I want to focus on data center first. Is there any way qualitatively or quantitative -- quantitatively you can help me understand the mix of different products embedded in the data center revenue. Specifically, I just want to see if most of your revenue is concentrated on the power solution or whether you have been able to diversify to include PDUs and more of a module or shelf solution.

Yes. So maybe, what we've said is, so far, all of our solutions have been rec level solutions. We really haven't been any more specific than that for competitive reasons and I think moving forward, we're looking at participating in all types of architectures. And so some of our participation may move out of the rack, but I think our primary focus will be on rack power in some of the peripherals.

So does that mean as power densification increase, you should be able to generate more dollars of revenue per rank or per watt.

Yes. I mean that's what it means because the challenges are greater, the higher the power levels. And that's what we've seen. Each generation brings us a bit more opportunity as a supplier of precision power.

Okay. Because obviously, it's outstanding to have doubled the revenue here last year, and now you're tracking to 50%. I'm just wondering if -- looking forward, there is growth acceleration on a year-over-year basis.

Yes, I think there is. I think looking back, we've tripled the size of this business in 2 years, which is impressive. But looking forward, we think there are 3 growth factors. Obviously, the first one is our business with the hyperscalers. We're well embedded there and well established with these customers and there are plenty of opportunities for us. The second one is the second wave I referred to earlier. These are customers that don't require a lot of engineering support. We could use existing technology blocks quickly meet their needs. And again, over the next couple of years, I think we can grow them to a size which is similar to our largest hyperscale customer today. And finally, since we're a technology company, really a technology leader, we see a lot of opportunity in 800 volts because this is a technology transition. And as this develops, I think it's a great opportunity for us to gain share and grow content.

Great. Okay. And then just one quick follow-up on the semi side. Your -- you guided the revenue in the second half would be up 50% year-over-year and implied guide for Q3 suggests to me that there is a sizable de-acceleration from Q3 to Q4. And I was on that kind of surprising because looking into the projects that your customers are working on, I would imagine there is a higher growth into the first half of '27, higher growth for your customers. So how come they are not prepared for that by building inventory, why should your revenue on the semi side to accelerate into Q4 when your customers are facing a step-up in shipment into the first half?

Yes. I think if you look at the numbers, Mehdi, it depends kind of how you attribute Q3 and Q4. I think we'll be -- what we said in our call is that we're shipping in line to demand. So that's how you should think about that.

Okay. So your customers are still not building inventory with in advance of complete system shipment.

Yes, it's hard for us to comment on our customers. But I think what we've said is that we believe we're shipping in line to their demand. We're not holding them up, and we're delivering what they're asking.

Our next question is from Jim Ricchiuti with Needham & Company.

A quick question just on Thailand. Did you say what kind of volume you anticipate in Q4. And as we think about the scale-up of the facility, I mean just given the strength you're seeing in both semi and data center, how are you thinking about allocating resources there to those markets?

Yes. Good question, Jim. So if you take our Q3 guidance, of $640 million in revenue. That implies that we already have more than $2.6 billion in revenue generating capacity in the company. And we continue to build capacity. So we're investing in existing factory sites, including adding new buildings. And that's going to allow us to continue to grow in Q4 and into '27, but all that's going to be augmented by our new Thailand factory.

And there's a lot of activity in Thailand right now, and we'll be producing first revenue there in Q4 of this year, so next quarter. And something notable is that we are qualifying big customers in Thailand. And these are big customers in the data center and the semiconductor market. And so we're leading with our premier customers and I think that's a real positive for the company.

But again, given what you're seeing in the market at this point, have you -- are you allocating more of the production capacity as you look into 2027 is semi or to data center? Or is that not changing versus your expectations when you started thinking about accelerating the capacity build?

Yes. I think we're basically investing to meet the capacity needs of both markets. And in addition, we're investing to meet the needs of the Industrial & Medical market. So I don't see any limitations in our capacity. We've invested heavily in CapEx for a number of years now. And we're using this capacity that we've built. And if you look at our investments, the bulk of the investment has been in data center in the past couple of years. But the return on investment and the payback is very quick on this CapEx for data center. So I think it's money well spent, and it's helping to fuel growth for the company.

Jim, one thing I'd just comment is, I think we said in our prepared remarks, we expect to see revenue growth into 2027. And we believe that we've delivered 2 straight years of over 20% growth. We could see over 20% growth next year as well. So some of that additional growth will come out of Thailand. We haven't quantified it specifically. And we haven't tied down how much each market. As Steve said, you should think about both data center and semi large customers going in there. But Thailand will certainly help support another growth year of over 20% for AE.

Our next question is from Scott Graham with Seaport Research.

Congratulations on a great quarter. I wanted to maybe understand a little bit some of the dynamics within semiconductor that has you at a market pace, whereas I guess I thought with some of the wafer intensity improvements relative to your plasma power solutions that maybe you'd be a little higher than that. Is there any way of size. I know that quarter-to-quarter is not a measure, but it's just sort of what I put out there to start the question, and that is would you guys be able to size for us the systems business versus the plasma power business right now and which one is growing faster?

Yes. We haven't broken it out specifically. But I think a couple of things we said we thought we could grow the system business by $40 million by 2030. I think we're well on track to do that. So that gives you a little bit of a sense. It's obviously much smaller than the overall plasma business. But you also have the service business, which is circa high teens or 20% of the plasma, which is growing much slower. That's right in our disclosures. So I think if you could take those couple of things and maybe you're looking at 25% or so of the semi business, is maybe not equivalent to WFE. So I could give you a little bit of a way to triangulate that growth relative to our numbers versus what you might hear in the market around system growth.

That makes sense, Paul. The other question I wanted to ask is you know I'm kind of a one that has a lot of interest in the I&M side. And I was just wondering one of the issues that the company was having until really the last several quarters was that you had design wins, but they were in a lot of different markets, and you kind of -- you've been looking for an acquisition, to kind of consolidate that business, give you some more critical mass. But I'm wondering if -- has any market maybe started to grown up over the last several quarters, where maybe there is some more critical mass that you're serving I mean, within medical, is there a market there? Or is that just more customer-centric. In industrial, is it automation. I'm just trying to understand how the dynamics of that business looks going forward even without an acquisition.

Yes. Yes. Thanks for the question. The I&M business has gotten a lot better in the past few quarters. So if I take a look at the bookings, they're almost double. If I look at the bookings per quarter the last 3 quarters, they're almost double what they were the previous 2 years. And so some of that is the general market recovery. This is the post-COVID hangover, the I&M market has been going through. But some of it is new product design wins ramping to volume. So we've been very pleased that some of these wins are actually going to high volume now. And so those 2 vectors are really helping push our I&M business upward. If I look at some of the distributor metrics, our bookings are up 80% year-on-year. Resales were up 40%. Our sales into the channel are up 45% and our inventory turns continue to improve. So I'd say the industrial medical market is definitely healthy for Advanced Energy right now, and we see strength in both Industrial & Medical.

I think we're seeing greater success in areas like test and measurement, aerospace and defense, factory automation, robotics and anything else related to AI. We tend to play in the high end of the market. And there's a lot of activity there right now. And a lot of our wins in the past 3 years are starting to bear fruit.

Our next question is from Elizabeth Sun with Citi.

I guess my first question is on the data center as well. Across all the hyperscaler customers, your position is very strong with one of them. I mean, very concentrated on one of the hyperscale customers. So I was just wondering, is your position strengthening over the rest of the hyperscale customers over the past course like, say, 1 or 2 quarters? And how do you balance your resources on continue supporting biggest hyperscaler customer growing your share in the others and support the ramp of the second wave customers.

Yes. We have said, Elizabeth, that there -- our resources are focused on a very select group of hyperscalers. And so for us, it's very important to be successful at the customers we selected. These hyperscalers are very engineering intensive. They expect a lot from us and they expect it quickly. And so we've been successful with that strategy. And we have to be careful not to dilute our resources and so that's why to augment our business with the selected hyperscalers, we've chosen to go with the second wave approach because the second wave customers aren't nearly as engineering intensive as the hyperscalers are.

So we can spin off derivatives using our existing technology blocks relatively quickly. And it doesn't strain our engineering resources. So that's our strategy, not necessarily to go and broaden our hyperscale exposure but to really take our technology and bring it to the second wave customers.

Understood. That's very helpful. My second question is on the semi side. I think, if I remember correctly, you were expecting some share gains in semi from the new products in the diaelectric etch in the second half of this year. So I was just wondering how that share gain -- how's the share gain progress is at this time right now.

Yes. We haven't been specific about our share gains this year. But what we have said is that we're ahead of plan. If you take a look at what we discussed at our Investor Day in December of '24. We're ahead of that plan as far as incremental revenue due to share gains. I think it's a combination of things. One is in our traditional areas of strength like plasma power. We're really focused on gaining share in conductor etch, dielectric and in deposition. And in the system power area where we really haven't emphasized this area in the past, we've been successful getting some major wins, both for process equipment as well as test equipment. And these are meaningful wins which taken together with what we're doing in plasma power should power a fair amount of share growth over the next 4 to 5 years.

A lot of what we're doing on the plasma power side is at the very leading edge. And that takes time to show up in our revenue, but it tends to be very robust, it tends to last a long time. And so that's why when I think about market share, I'm thinking about what are we going to do over the next 3 to 5 years.

Our next question is from Duksan Jang with Bank of America.

Just a follow-up on the semi's question, and you've been consistently talking about WFE outgrowth. Your customers have been pretty strong. You've been talking about your own product ramps. I think the market share gains that you just talked about. Putting all of these together, what kind of visibility or how much confidence do you have in actually outgrowing WFE for the foreseeable future and I'm asking just because clearly, you've been gaining share relative to your competitor, but because deposition and edge haven't been really growing in line with WFE. I think you've undergrown the market the last couple of years. So obviously, you're doing great with the results today, but I'd like to hear any more visibility color.

Yes, it's a good question, Duksan. I mean, our overall goal is to outgrow WFE. Obviously, the mix of WFE matters, which segments are growing, at what pace and what our participation is in them. But as we look at the market, as Steve said earlier, we highlighted $140 million. We built a share gain that we could accomplish. This was from our 2018 Analyst Day -- our 2024 Analyst Day. We think we're on track to beat that. Now WFE is bigger, so we'll have to see what the pace of growth is. And certainly, the -- again, the mix of WFE growth over time will have an impact. But when we look at our fundamental systems business that's within our semi market, we still feel very good about our ability to outgrow WFE.

Got it. And then one on more gross margin outlook. I think people have already asked around capacity and your product mix. But I think in the past, you've mentioned volume is also a pretty big driver. For your target model, I believe it was $3 billion sales and 53% -- or 43%, sorry, gross margin. We're kind of getting in that range in a few years. So as we expand further with the Thailand facility and clearly, the end markets are really growing strong. What kind of gross margin, I guess, puts and takes and further opportunity you have ahead?

Yes, it's a good question. You're right about volume, and we've certainly realized some of that benefit. We've also realized a lot of the benefit from our factory consolidations. The areas where there's more room to improve gross margins, if you will, relative to volume is in the areas of product mix and now a little bit of manufacturing efficiency. When we talked about product mix, we talked about new products and maybe becoming a larger portion of the total over time. That takes a little bit of time. The fact is we're running way ahead of the revenue number based on volume. So we think from these levels, there's still a fair amount of improvement that we can gain from that mix of new product revenue.

And like I said in our prepared comments, we're already seeing that start to happen. So that will be the #1 driver of I'll say, faster margin improvement than just volume is from that ongoing improvement from product mix. And similarly, I think we can work out additional benefits from manufacturing efficiency. We are scaling rapidly. I think we're getting better at that over time, and we'll see those benefits fall through as well. So on balance, we're fortunate to be running quite a bit ahead of our 2024 model on revenue. That's contributing to higher gross margins in aggregate. But certainly, we think we can go beyond our model because, in fact, we're starting to exceed the revenue side. So thinking forward, there's no reason we shouldn't be able to deliver more than 43% margin as margins improve or as volume increases and the new product mix continues to strengthen.

Our next question is from Quinn Fredrickson with Baird.

First question just on 800 volt. Just wondering if you could talk about how you feel you're competitively positioned as the market makes the transition and why hyperscaler might choose to buy your solution instead of an integrated sidecar that some of your competitors might make?

Yes. Our positioning 800 volts is very similar to our current positioning. We're a technology leader. And so what we've chosen to do at 800 volts is to create a number of different modular solutions. And this gives our customers a lot of flexibility on how they mix and match our technology. And this has been very well received, this approach. And so I think we're competitively positioned pretty well, actually, based on the feedback we're getting from customers. They also know that we can ramp to volume. We've been doing it now for the last few years in data center very successfully. And so what we offer customers is a mix of technology leadership and manufacturing muscle. That's a good combination.

You discussed potentially north of 20% growth next year and talked about semi and data center side. What does your visibility to the industrial medical side look like in the next year from a new design win standpoint just underlying market?

We don't have the same visibility necessarily as we do in semiconductor and data center because those are markets where it's very large customers, there's just a few of them. But in I&M, I really need to look at the trends and what we see is a very positive trend in the past 3 quarters. And we think that trend will extend into '27 and probably through the entire year. So our expectation is that I&M is going to do very well in 2027. We're well positioned to take advantage of that given our design win pipeline and our capacity. So I think we're looking at a very strong year in 2027, as Paul said, up at least 20%, and that will be up in all of our markets.

Our last question will be from Daniel Taglio with Stifel.

This is Daniella. I'm on for Brian Chin from Stifel. My first question is on semi. I know you have touched on the second half, 50% year-over-year growth. I was wondering if you could maybe tell us a little bit more? Is that coming from your existing flagship products or your new products ramping faster than expected?

Daniella, most of that is from the existing products, but we also see contribution from the new products, but it's not as significant this year as it will be next year. We have good visibility into next year from our customer base in semiconductor and so we think it's going to be a very good year. And it's going to be a good year both in plasma power, where we see a number of our new products becoming more significant on the revenue line as well as with our existing products. So yes, I think we're well positioned to grow both in the second half of this year into 2027.

Great. And then if I could touch on gross margin a little bit more also on semi cap. If you have a shift in mix towards semi, does that enhance your ability to drive gross margin leverage and maybe to quantify that a little bit for every 100 basis points. Could you see that on a $50 million quarterly revenue increase?

Yes. So maybe 2 parts to that answer. We've commented before that semiconductors on balance have the better margins than the rest of the company. So yes, we had more mix shift towards semi on balance, that would be positive. But I think as data centers improved, we've talked about that, that sort of closed the gap and is approaching corporate average. So it's not as big as a gap as it's been historically. The second thing is we do continue to get volume leverage. We said for every $50 million of quarterly revenue. If you go back to when we put the model out, that was worth 100 basis points. But as you know, as your revenue grows, it's the same dollar contribution, but the percentage impact is smaller because the denominator is bigger. So at this level of revenue, it's about 40 basis points for every $50 million of quarterly revenue. That's the same dollar contribution for that move. It's just a smaller percentage because the revenue is much higher.

Thank you. This concludes our conference call for today. You may disconnect your lines at this time. Thank you again for your participation.

[Audio Gap] Energy team. We have Steve Kelley, President and Chief Executive Officer; Paul Oldham, Executive Vice President and Chief Financial Officer. And then in the audience, we also have Edwin Mok, Senior Vice President of Strategic Marketing and Investor Relations. Before we begin, Paul had some -- a couple of comments to make.

Yes. Just a reminder that our comments today are subject to a number of disparate risk factors, and we encourage you to have a look at those in our filings.

Awesome. Crazy week so far. I'm sure there's a lot going on at Advanced Energy as well. Steve, could you start us off maybe by giving us an overview of what you've been seeing in the first quarter and then in your second quarter outlook?

Yes. So first quarter was a good quarter for us. We came in above our midpoint guidance in both revenue and profitability. We also crossed the gross margin threshold of 40%. So we've been trying to get above 40% for the 5 years I've been CEO. So it was nice to finally make that move. And we think we can maintain the move above 40% and keep it there for the remainder of this year. And our ultimate goal is to get to 43%. The 40% was a big milestone.

We also announced that we saw a large increase in orders from our semiconductor customers in the first quarter. And so we have a strong backlog now for the remainder of this year into '27. And from what we hear, that semiconductor strength should last into '28. So that was a good quarter for semiconductor orders. In data center, we were able to start sampling our 800-volt solutions. And today, we actually announced some more specifics on those solutions.

And so we've been sampling a number of customers over the past 3 months and getting really good feedback on our modular solutions, which feature very high power density, best-in-class power density and very high efficiency. In Industrial and Medical, we talked about the health of that market. It had been in a correction mode for the better part of 2 years. And we see industrial medical normalizing now. Backlog is strengthening, and we expect our revenue to increase every quarter this year. So bottom line, all 3 markets are looking very strong in 2026 and into '27.

Got it. We'll certainly get into each of those segments for sure. But when we look at the high level and look at the full year, you've certainly also raised the guide for this year. Maybe walk us through which segments you're seeing the most strength in and then maybe perhaps some weaknesses in others.

Yes. So we don't see any weakness. So what we've said is in semiconductor, it's very strong, and we're expecting greater than 20% growth this quarter year-on-year. And the second half is going to be more than 30% better than second half of '25. In fact, the demand is so good in semi that we're going to open our Thailand factory a quarter early. So we'll be able to start producing semiconductor products in Thailand in Q4 of this year.

In data center, demand is very strong. What we've seen is that there's been some movement in the forecast, and it really is tied to the availability of memory modules and logic silicon basically. And so as we enter each quarter, there's a lot of uncertainty about the exact mix, but we know the overall demand is excellent. And so what we see there is a very strong second half. And again, the other factor behind our Thailand factory opening is data center. So in addition to the semiconductor products we're going to produce in Thailand, we're also producing data center products.

And that's important because data center is high volume, low mix, and it's a great way to start a factory, right, because you absorb some of those initial fixed costs that you have in any new factory start-up. So demand in semis, data center, very strong. In I&M, it's also looking good. We're seeing a lot of our design wins transition into production. And we're seeing the health of our distributor channel is much better than it was a year ago. So we've seen our inventories in distribution go down 6 straight quarters in a row, and it's pretty close to equilibrium levels. So we're pretty optimistic about our I&M business moving forward.

Appreciate it. Going into semiconductors, as you say, we can probably talk about both demand and supply. But starting with the demand side, I think it's been pretty clear that a lot of the semicap vendors have been raising their guide for this year and next year. Clearly, a lot of demand signals here and there.

I think consensus at this point is the WFE market rises maybe 30% this year, another 20% to 25% next year. First of all, is this similar to what you're seeing? And then secondly, clearly, you have more exposed side to depo and etch. So how should we think about AE's growth in relation to the WFE market?

Yes. So I think we're going to have the best year we've ever had this year, and we're seeing very strong growth. As I mentioned before, second half is going to be at least 30% better than last year second half. And there's upside to that. And again, that's why we're opening our new production line in Thailand. I think for us, the more important question is, are we positioned to gain market share over time? And we are.

So we had introduced our new products, eVerest, eVoS, the NavX back in mid-'23. So we've been working closely with our customers over the past 3 years, getting these products designed into leading-edge processes. And not just logic but also memory processes. And we've been successful. And so we're starting to see some of that revenue appear in our revenue line, but we'll see more of that in '27 and it's going to grow in '28, '29.

So we feel very good about our ability to grow market share in semiconductor. In addition to that, referring to WFE, I think it's well known that at these advanced nodes, there are many more etch and deposition steps. So the etch and deposition intensity goes up. That benefits our customers and it benefits us.

On these new products, you've been mentioning them for quite some time. Obviously -- is there any way to think about maybe units and content? How are they different from the prior products? Do they give you any advantages in certain areas?

Yes, that's a good question. So these new products are replacing, I would say, pretty mature products introduced more than 10 years ago. And so the value that we're providing is much greater than what was provided by the old products, which means it's a higher price. It's also higher margin. And so that's an accelerator for us as these leading-edge processes ramp the volume in memory and in logic, we get the benefit of share gain, higher prices and etch and dep intensity.

Could you also maybe talk about the design wins that you might have had? And how long do these wins typically last for? Is it a 2- to 3-year time line or a 3- to 4-year time line?

Yes. So I think on the design wins, it's a multiyear effort. So the way it works is, we introduce our platform. The customer evaluates the platform and their laboratories comes back to us with requests for changes. Then we create a derivative product and they work on it some more. We go back and forth and then our customer brings their customer into the mix.

And so that's where we're at with most of these products now, where our products are being tested in end customer wafer fabs essentially. This is a multiyear effort. But the good news is it's a sole-source design win. There's no competition once we win the slot, and it lasts for years. So particularly for some of these nodes that will last for 5 or 10 years, they'll keep buying products from us. And after that, they'll continue to service products with us. So these are good revenue annuities for us, these design wins in semiconductor.

Is there any way to quantify the ramp? You said it starts in '27 and then it runs through '29 plus. Clearly, you're seeing semi strength beginning probably next quarter and then in the second half. I'm sure all of this is incremental to what you're seeing today. So maybe help us quantify this benefit over the medium to long term.

Yes. What we said was in '25, we have new product revenue in semiconductor between $10 million and $20 million and that we would at least double that this year. We haven't quantified beyond that. But these ramps tend to ramp pretty steeply. So once the process is qualified, you see a pretty big increase in orders. So we're expecting a significant ramp in '27 and in '28 for both conductor etch in '27 and dielectric in '28.

Got it. Are these in any ways, cannibalistic to your old products? I think you said these are very mature 10-year type products that you're replacing. But is it -- do you think pretty much all majority incremental or any others?

I think it's all incremental. The good part here is that we are providing something that's very difficult for our competition to compete with. So this is increasing our market share at the leading edge. And then what we're seeing is once we've won that leading edge node, in some cases, the customers are trying to basically back convert. So they'll go a little bit nearer in time to see if they could use our technology for more mature processes. So there's also -- I guess, I'd characterize it as an upgrade opportunity in some cases that we haven't quantified yet, but it could be substantial.

Got it. One more on the semis market. So increasingly, obviously, the industry is very supply constrained today. We're seeing increasingly more fab upgrades, say, rather than greenfield sites. So how does that have an impact if it has any on Advanced Energy's tools and services?

Yes. I think we've heard a lot of talk about upgrades in the NAND fab business. It really doesn't impact us very much because our penetration in NAND fabs is less than what we have in DRAM and advanced logic. But I think where we'll be able to gain share in NAND is in the greenfield fabs, where I think the end customer has an opportunity to upgrade to systems that use our latest solutions.

Got it. And speaking of your customers, I guess, is there any lag between when your customers start to ship orders and then when you maybe recognize revenue? Is it usually before your customers? How does that typically work?

Yes, it's generally before our customers because we have a subsystem that they plug into the larger system. And the way we operate with our largest customers is in a just-in-time inventory mode. So they'll have bins where they keep a certain inventory of our products, and we'll ship to the bin level essentially. So that prevents them from running out of product.

Got it. On to data center. Clearly, one of the fastest-growing segments for you. I think you've had 8 record quarters of straight growth. This second quarter, I think you said the growth may be moderating a little bit. Could you explain what that could mean and what you're seeing for the rest of the year as well?

Yes. I think the second quarter forecast is really a short-term phenomenon. It's not an indicator that demand is going down for us. It's just an indication that some of our customers were having difficulty securing some products. So there are downstream supply chain issues that our customers are dealing with. And I said on the call that I thought that our customers would be able to solve most of these issues over time. And we are seeing that. I also said there's upside bias to our numbers in data center. And so I think we guided to a relatively conservative number, a number we knew we could hit, but we're certainly aiming for much better.

I see. And I've been comparing your semi market to the WFE market. I think a lot of investors probably compare your data center market to the cloud CapEx. I know it's not exactly one-to-one, but what's the relationship there? And perhaps what could help you upside or downside the headline cloud CapEx number?

Yes. I would say the relationship is loose at best. So if you take our -- if you take a look at our growth last year, it was 107%, I think, '25 versus '24. So, we outgrew the market by a wide margin. This year, our forecast shows that we're going to grow at least 35%. So it's difficult to correlate with CapEx spend.

But what we think based on what we see in front of us is that we're gaining share with our customers. Our margins are getting better, and we're filling our factories with these products. And it's another reason we're opening Thailand in Q4 is because we run out of capacity for data center in our existing factory network. So the business is very healthy.

I see. Then are there any other metrics that investors should be aware of? I think some people are trying to track the gigawatts that are installed. Is it the power architecture type? Or what are some helpful metrics that we should be looking at then?

It's a tough one because our strategy in data center has always been to focus on the hard power problems to solve those problems and get paid for solving them. And so we're not trying to address every opportunity out there. We're very selective. So I think you're going to see years where we have outsized growth, other years where we have more modest growth, but it's kind of independent of the large metrics.

Got it. For 2026, specifically, you said you have probably more upside than the 35% that you mentioned. Then would the gating factor be more supply or demand? Because the demand seems to be almost out of control. So any thoughts around that?

Yes. I think the gating factor is definitely supply. So it's downstream supply chain constraints. And if our customers are successful in addressing those constraints, yes, I mean, the demand is quite impressive.

Got it. On your specific customers, I know you can't talk specific names, but you do have exposure at all 4 hyperscalers. Are there any differences in what you ship to each of these customers? What's your general relationship there and your exposure?

Yes. It's interesting because every customer wants something different. So that's our kind of market. Our business is bringing best-in-class technology to our customers and then customizing that technology to their needs. And so that's a characteristic of the hyperscale business as everybody wants something a little bit different. And we're able to accommodate that for a limited number of hyperscalers.

As we look forward, we have a second wave of customers coming, basically non-hyperscale customers who want similar products, but they're not as engineering intensive. So, we take existing technology blocks that we've already developed for other purposes, put them together, add a few bells and whistles, and we've got a product for the second wave customers. And it doesn't drain resources from our main focus, which remains the hyperscale.

Got it. When you say second wave, are these mostly neoclouds or enterprises? Help us understand that nomenclature. And I think you've also said that these are more for '27 and beyond. Is that still the right outlook to have?

Yes. We use the second wave nomenclature to refer to anybody who's not a hyperscale customer. So it includes neocloud, it includes enterprise customers. And we have none of that revenue in our forecast for 2026. So we believe it begins in earnest next year. But there's a possibility we could start to ship to those customers in the last quarter of '26, but it's not in our forecast yet.

Got it. Your hyperscalers, you've been mentioning they're sole sourced. Is it also similar for these second wave customers?

It's a mix like it is with the hyperscale, mostly sole-source, but also somewhere we're 1 of 2 sources.

Got it. Does that have any impact to your margins or pricing, do you think?

No. No, we think the business for the second wave customers should come in with very similar margins to what we get for the hyperscale products.

Understood. I do want to dig into the competitive landscape. So people obviously talk a lot about the competition. Delta is also doing pretty well. So could you talk about what makes you differentiated? Do you think you can gain further market share at existing customers on top of these new customers?

Yes. So I think our position in the market is we're the technology leader. And I think we have one peer competitor it's Delta. And so our objective is to leverage our technology advantages. And together with that, we combine speed of development as well as our ability to ramp to volume quickly.

So that combination has been a winning combination for us. So our willingness to invest a lot of money in our factories, our willingness to invest a lot of money in our engineering team and our focus on highest power density, highest efficiency and highest reliability. That's a winner in the data center market.

Is there any way to think about the relative market share today?

We don't measure market share in data center. So we're focused on a high-quality business that has a good return on investment and good gross margins and less concerned about market share.

Understood. You said the gating factor is probably going to be supplies. You've mentioned the Thailand factory. But when you add all of those together, I think at some point, you said you can satisfy about $3.5 billion of sales. Well, here we are, the WFE market is exceptionally strong. Data center is exceptionally strong. Industrial is likely recovering. Is it going to be enough? And do you have any plans further out than the Thailand fab?

Yes. So I think what those figures I discussed in the last call, the $3.5 billion in total. Those are kind of our nameplate capacity, but we have the ability to stretch that capacity to probably $4.5 billion, $5 billion, if need be. So I don't think capacity is going to be an issue for the company anytime soon. The other thing to keep in mind is, we'll likely do at least one acquisition, which will come with factories. So yes, I think we're okay on capacity.

Got it. More on the 800 volts, clearly a big tailwind for the industry. What's AE exactly doing in this architecture? How are you guys differentiated? Any implications to pricing, margin, share? Any of that would be helpful.

Yes. So for us, it's the technological step. I think we've gone from 12 volts to 48 volts and now we're going to 800 volts. And the nice part about this is it leverages the same things we've done before. And so this focus on high power density, high efficiency is the same thing we're bringing to 800 volts. We announced some new products this morning in our press release. And basically, we're claiming best-in-class power density and efficiency.

And these aren't future products. These are today's products. So we've been actually working with customers for months now with these modules, and we're getting some very good feedback. And you'll note that it's a modular approach. And so that allows the customer to customize to their needs. So we can build modules. We can put the modules on a board for the customer. We could construct heat sinks, the whole thing. But it leaves the ultimate solution in the hands of the customer, and they can work with us to arrange the parts.

Then should we assume you're designed into each specific customer racks rather than maybe say NVIDIA or some specific accelerators?

Yes. So we're focused just on the rack. So this will be taking 800 volts in and then down converting it to whatever voltage customer needs.

Got it. When we think about the 800-volt architecture, are these usually greenfield data centers? Do obviously -- the hyperscalers, do they need to invest in completely new sites to support these? Or are they also upgrades type revenue as well?

To the best of my knowledge, it's mostly greenfield. There might be some upgrade plans out there, too, that I'm not aware of. But that's the reason why we don't really see meaningful revenue in 800 volts until the second half next year. But really, 2028, I think we'll see.

Got it. Any implications on content and margins from the 800 volts?

Yes. We've taken a look at that, and we think our content goes up with the 800-volt transition. We also think our margins get better. It's a different type of product, right? It's modules and boards and not boxes. And so it's easier to manufacture, and I think we can drive better margins in that business. So we're looking forward to becoming a major supplier to 800-volt data center solutions. I think we're in a good position from a technology standpoint, and we certainly know how to build these types of modules.

Got it. A little bit on the industrial and medical market as well. A lot of the -- at least in the semiconductor market, we've been seeing some recovery, particularly in the industrial side. So help us understand what you're seeing in the market and what kind of outlook you have for the foreseeable future?

Yes. So maybe going back a couple of years, I think the industrial and medical market was the last market to catch up after the COVID supply chain shortages. And that catch-up occurred in 2023, a large part. And so after that, basically, industrial medical distributors and customers had too much inventory. They were out of phase with the inventory and the demand. So it took about 2 years to work through their inventory problems as a market.

And now we're in a situation where inventories are where they need to be and the customers are ramping new designs because a lot of our customers are focusing on trying to consume the inventory and selling older products. Now they're focused on the future. They're ramping up new products, and that's causing a lot of our design wins to ramp to production. And so in addition to a market recovery, we're also seeing our design wins kick in, and that helps us outgrow the market and also helps our margins because the new products have better margins than the old product.

Just one actually quick question going back to data center. How should we think about your visibility? What gives you confidence, maybe, say, each specific quarter and maybe the next couple of quarters? Is that your customers really just giving you that visibility?

Yes. Our customers have given us good visibility, I would say, most customers at least 12 months right now, and they're giving us a lot of positive signals for '27 and '28. So the primary discussion point I have today with customers is our ability to ramp to their needs. And I think they've been satisfied with the answers. We have the factories that we can use to build their products.

And also, I think we've got a good supply chain strategy where we're staying ahead of the demand. And we're putting in place more inventory, quite frankly. And this is allowing us to be more flexible as the demand in data center shifts based on what our customers can procure in the way of memory and logic chips. And so we've been able to keep up with their demand, and that's a big plus.

Got it. And then I do want to leave some time for Paul as well. Gross margin is clearly a very big achievement and milestone, as you said. What inning are we in at this point? You've come a long way. Your goal is 43%. Help us with any color.

Yes, it's a good question. First, thanks for the comment because I think it is a big accomplishment what we've done to date. But we think we're still in kind of the middle innings here. As you mentioned, we crossed 40% in Q1. We expect Q2 to be up again, 20 to 50 basis points. And we said in our call that we expect to exit the year around 41%, but our goal is 43% plus. And we think we're well on our way to that.

The biggest contributor to that will be as our new products start to layer in, they come with substantially better margins. That will add 200 to 300 basis points to the margin profile as those products become more substantial. The good news there is we're already seeing that start to have an effect. Even in Q1, we started to see that benefit. We'll get the benefit of additional volumes, and there's still some opportunities to bring out some improvements in manufacturing. So we're pretty excited about the margin opportunity for the company looking forward.

Got it. As data center continues to ramp, this segment historically has been slightly dilutive. How should we think about it going forward?

I think a couple of things. One, we've made a lot of progress to bring gross margins up. As Steve commented, our new products here can also help with the margin as the content and the complexity increases. In addition, as the volumes in these increase, we get some pretty significant manufacturing efficiency on those, which maybe doesn't contribute to the product margin, but does contribute to the overall company margins. So we think across our markets, we can continue to improve margins here and continue to make progress in data center as well.

Besides gross margin, how should we think about your OpEx trajectory? Clearly, a lot of investments do need to be made. When we think about the EPS leverage, what would be the biggest contributor?

Yes. Looking at OpEx, we actually held OpEx flat or a little bit down in Q1. I think that was timing with year-end. We do expect to see OpEx continue to grow throughout the year to get to about $460 million for the full year. But the thing to remember there is our goal for OpEx is to grow about half the rate of the revenue growth. And based on our guidance, we would be about on that track. Now last year, that was much less than half. And I think as revenue accelerates, we can repeat that performance again.

Bottom line is we don't have to make major incremental investments. We're going to continue to double down on R&D because that's important. It's the lifeblood of the company. We can continue to be more efficient and leverage fixed costs in our infrastructure and SG&A. And overall, that should allow us to continue to hit our leverage targets. Overall, for the company, we should be able to drop 35% to 45% of every incremental revenue dollar towards the operating income line.

Understood. Perhaps the last one is, when we think about cash returns, you've mentioned acquisitions. You recently also had a convertible note offering. What would be the priorities in cash spending and then perhaps any return projections as well?

Sure. First, you're right. We did just raise additional money. That was primarily to recapitalize or refinance our existing convertible notes. So we got much better terms on this one. So we've lowered our cost of capital in that regard. From a cash perspective, we're clearly investing heavily internally in our data center and semiconductor markets, both in terms of engineering as well as in CapEx to build the factories capability and capacity. And with respect to external spending, we would allocate that towards smart M&A, largely in Industrial and Medical market.

Sounds good. Well, thank you, Steve and Paul.

Thanks very much.

Greetings, and welcome to the Advanced Energy First Quarter 2026 Earnings Call. [Operator Instructions] As a reminder, this conference is being recorded. It is now my pleasure to introduce Edwin Mok, Senior Vice President of Strategic Marketing and Investor Relations. Please go ahead.

Thank you, operator. Good afternoon, everyone. Welcome to Advanced Energy First Quarter 2026 Earnings Conference Call. With me today are Steve Kelley, our President and CEO; and and Paul Oldham, our Executive Vice President and CFO. You can find today's press release and presentation on our website at ir.advancedenergy.com. .

Before we begin, let me remind you that today's call contains forward-looking statements that are subject to risks and uncertainties that could cause actual results to differ materially and are not guarantees of future performance. Information concerning these risks can be found in our SEC filings. All forward-looking statements are based on management's estimates as of today, May 4, 2026, and the company assumes no obligation to update them.

Any targets beyond the current quarter presented today should not be interpreted as guidance. On today's call, our financial results are presented on a non-GAAP financial basis unless otherwise specified. Detailed reconciliation between our GAAP and non-GAAP results can be found in today's press release. With that, let me pass the call to our President and CEO, Steve Kelley.

Thanks, Edwin. Good afternoon, everyone and thanks for joining the call. First quarter revenue came in above the midpoint of guidance, driven by record data center revenue. Total revenue increased 26% year-on-year and gross margin exceeded 40%. In the second quarter, we expect to deliver record revenue, largely due to strength in semiconductor. .

Looking into the second half of 2026, we see increased demand in all of our markets. We are particularly well positioned to benefit from AI-related capacity investments in data centers and wafer fabs. We are also seeing steady improvement in the industrial medical market as evidenced by a 14% sequential increase in bookings, and a growing backlog.

We delivered over 40% gross margin in the first quarter. The culmination of a multiyear effort to improve our manufacturing and efficiency and product differentiation. Our investments in leadership technology and world-class manufacturing are paying off. Looking forward, we believe that we can further increase gross margin as high-value products ramped to volume and manufacturing efficiency continues to improve.

Given our progress over the last few years, we are confident that we can achieve the longer-term goal of greater than 43% gross margin. Given the strong demand environment, we are executing our capacity expansion plans in Malaysia, the Philippines and Mexico. Moving forward, we will focus on building out capacity at our new 500,000 square foot facility in Thailand.

Qualification builds for semiconductor and data center products are kicking off this quarter, with initial production slated for late '26 or early '27. Exiting the year, we expect to have over $2.5 billion in revenue generating capacity. The addition of Thailand will bring total capacity to over $3.5 billion once it is fully built out.

Now let me provide some color on each of our markets. Semiconductor revenue increased quarter-over-quarter and was flattish year-on-year. In the first quarter, customer forecast strengthened considerably which we believe will drive record performance in 2026 and continued growth in 2027.

We are delighted by the widespread customer acceptance of our EBOs Everest and NavX plasma power technologies. These technologies enable significant improvements in throughput and yield at the leading edge and are expected to drive market share gains into the next decade. In addition, we are seeing wider adoption of these technologies across multiple generations of processes, and device types.

We are also benefiting from an uptick in demand for our system power products, largely due to recent wins in test and wafer fab equipment applications.

In data center computing, we delivered record revenue in the first quarter. Overall demand in the data center market remains very strong. Based on customer forecast, we expect second half revenue to be stronger than first half. We continue to make solid progress developing next-generation technology, including 800-volt solutions. We are working closely with multiple customers who view AE as a technology leader in this space.

The attributes, which have fueled our success in the data center market, power density, efficiency, reliability and development speed will be equally critical to our success in next-generation platforms. In the first quarter, leveraging our technology expertise and product portfolio, we secured multiple new wins with second wave data center customers.

Factory qualifications should be completed this year ahead of production ramps in 2027. Industrial medical revenue was up year-on-year but down sequentially. Although demand is improving, factory priorities in the first quarter limited our output. We expect to increase our factory output in the short term, which should enable I&M revenue to track bookings moving forward.

In medical, we secured multiple wins in therapeutic, diagnostic and life science applications. In industrial, we won key designs in test and measurement, factory automation and battery backup applications. We secured many of these wins by adding custom features to best-in-class technology platforms. enabling us to meet customers' unique requirements.

We have won a number of opportunities with new customers, many of whom discovered AE products on our website. Some of these wins have been quite large reinforcing our view that the new website is acting as a force multiplier in the I&M space. Telecom and networking revenue grew to its highest level since 2023, driven by the production ramp of several AI-related wins in the networking space.

Now I'd like to provide an update to our 2026 view. Based on strengthening demand and new product momentum, we are now expecting year-on-year revenue growth in the low to mid-20% range. This outlook represents the second consecutive year of greater than 20% growth for Advanced Energy. In semiconductor, we expect demand to start accelerating in the second quarter, supporting a stronger outlook for 2026.

With some of our new products moving into high-volume production later this year, we believe that we are well positioned to drive further growth in 2027 and beyond. In data center, based on strong customer adoption of our high-power AI solutions, we are raising our full year revenue growth expectation to the mid-30% range.

In the industrial medical market, we expect sequential revenue growth over the next few quarters, supported by improved market conditions and their production ramps of several key design wins. Now for some closing thoughts. First, demand across all of our markets is strong, and we are raising our growth target for the year.

While supply and cost challenges have begun to surface, we are well prepared to navigate a dynamic environment. Second, we continue to see strong pull for our new products across all target markets. Our design win pipeline is growing and is expected to drive higher revenue and profits in the coming years.

Third, we're proud to have achieved 40% gross margin in the first quarter, but we are not done. We have line of sight to 43% and based on the success of our new products and efficiency gains.

Finally, we have a solid pipeline of potential acquisitions and we'll continue to actively pursue opportunities, which make strategic and financial sense. Paul will now provide more detailed financial information.

Thank you, Steve, and good afternoon, everyone. Overall, we executed well in the first quarter. Revenue of $511 million increased 26% year-over-year and was ahead of our guidance driven by strong data center computing revenue. Importantly, we achieved our initial milestone of gross margins of over 40% despite ongoing tariff expenses and less favorable market mix than we originally modeled. .

It is the highest level since the Artesyn acquisition in 2019, highlighting the structural improvements we've made in operational efficiency and our product portfolio. With solid operating leverage, we delivered record operating income of $98 million. As a result, first quarter earnings per share were $2.09, exceeding our guidance and up 70% year-over-year.

Now let's review our first quarter financial results in more detail. First quarter semiconductor revenue of $219 million grew 4% sequentially, finishing just below our mid-cycle peak last year. Looking forward, our outlook is increasing based on stronger customer demand.

Data center computing revenue was another record at $194 million, up 9% sequentially and and 102% year-over-year. While demand remains high, we continue to experience frequent customer changes in demand mix due to various downstream constraints. While we expect this demand volatility to limit revenue in Q2, we anticipate the ramp of several programs to support a stronger second half.

Industrial medical market revenue was $72 million, down 8% from last quarter, but up 12% from last year. We prioritized factory production to meet demand in other markets, impacting revenue for the quarter. On the other hand, demand is strengthening as bookings grew 14% sequentially, reaching the highest level since 2023. Distributor sell-through increased again and inventory levels further normalized.

Telecom and networking revenue increased 17% sequentially and 16% year-over-year to $25 million, ahead of expectations due to strength in AI-related networking programs. First quarter gross margin was 40.1%, up 40 basis points from last quarter and 220 basis points from last year. Gross margin was above our previous guidance, driven by better product mix and lower other cost of sales.

Looking ahead, we expect to further expand gross margins on ramp-up of higher-margin new products, improved manufacturing efficiency and higher volume. Operating expenses of $107 million were down slightly from last quarter and at the low end of our target range.

OpEx increased 9% year-over-year, well below half of our revenue growth rate of 26%. As a result, first quarter operating income reached $98 million and operating margin was 19.1%, up 560 basis points from last year. Depreciations was $10.5 million, and our adjusted EBITDA was $108 million, up 66% year-over-year and also a record.

Other income was roughly breakeven versus $1 million in Q4 and mainly due to higher realized FX losses. For Q1, our non-GAAP tax rate was 14.5%, below our target, mainly due to timing of discrete tax items. First quarter earnings were $2.09 per share compared to $1.94 in the previous quarter and $1.23 a year ago.

Turning now to the balance sheet. Total cash and cash equivalents at the end of the first quarter was $700 million with net cash of $131 million. During the quarter, we increased inventory by $48 million, mostly in critical piece parts to support growth and improve supply resiliency. As a result, inventory days increased 10 days to 135 with terms of about 2.7x.

Correspondingly, DPO increased from 68 days in Q4 to 80% in Q1. DSO increased 6 days to 66 days in Q1 on higher revenue. As a result of the increased trade net working capital to support growth and the seasonal factors such as timing of incentive and tax payments, cash flow from continuing operations was an outflow of $6 million.

During the first quarter, we spent $37 million in CapEx as we continue to invest in capacity and capability across our factory network. We paid $3.8 million in dividends, and we repurchased of common stock at an average price of $209.6 per share.

Turning now to our guidance. We are forecasting our second quarter revenue to be approximately $540 million, plus or minus $20 million. We expect the majority of the sequential growth come from the semiconductor and industrial and medical markets while data center will moderate sequentially based on timing of customer deliveries. We expect Q2 gross margin to improve 20 to 50 basis points sequentially, driven by higher volumes and more favorable mix. We expect Q2 operating expenses to increase to $112 million to $114 million due primarily to investments in new products and annual merit increases.

We expect other income to be approximately $1 million and the tax rate to remain within the 16% to 17% range. As a result, we expect Q2 non-GAAP earnings per share to be $2.18 and plus or minus $0.25 on 40.6 million shares outstanding.

For the full year 2026, we are raising our revenue growth target from the high teens to the low to mid-20s. The increased growth outlook contemplates solid customer demand as well as some tightening in supply and increasing input costs. In semiconductor, we expect revenue to accelerate in the second half with 2H revenues likely up over 30% from the prior year.

In data center, despite a moderating Q2, we expect sequential growth in the second half and are raising our full year revenue growth outlook from over 30% and to the mid-30s. In Industrial and Medical, we expect revenue growth throughout the year on higher demand and increased factory output.

With continued improvement in gross margin and operating leverage in our model, we expect earnings to grow meaningfully faster than revenue for the year. Finally, we expect our 2026 CapEx will be in the $170 million to $180 million range up slightly from our previous outlook based on initial investments in the Thailand factory to support earlier customer qualifications.

Despite higher capital spending, we are targeting 2026 free cash flow to be at or above '25 levels. Before opening it up for questions, I want to highlight a few points. Demand is strengthening across our markets. Our diversification strategy is paying dividends as we are benefiting from accelerating growth in semiconductor, increasing investments in data center and AI infrastructure and a recovering industrial and medical market.

In addition to positive market trends, we expect our design win pipeline to contribute incremental revenue in '26 and to support more meaningful growth in 2027 and beyond. We are excited to have achieved gross margin of over 40% this quarter and expect to further improve our margins for the full year.

Longer term, with higher value new products, ongoing improvements in factory efficiency, and higher volumes, we remain confident in our ability to achieve our long-term goal of over 43%. Finally, our balance sheet remains strong, enabling us to invest in capability and capacity to capture growth ahead while providing ample liquidity to pursue strategic acquisitions that create shareholder value. With that, we'll now take your questions. Operator?

We'll now be conducting a question-and-answer session. [Operator Instructions] Thank you. Our first question is from Steve Barger with KeyBanc Capital Markets.

This is Jacob Moore on for Steve, actually. First, I was hoping you could provide us some more detail on the uptake of the new semiconductor products. I mean, for leading edge, of course, but also for the opportunity at nodes larger than 2 nanometers that I think you were alluding to.

How has qualification and uptake progressed since your last update how do you expect it to progress through the rest of the year? And what milestones are you watching out for to determine different levels of success in that rollout?

Jacob, this is Steve. Yes, we're seeing quite a bit of uptake on our leading-edge technologies, namely Everest, EVOS and NAVEX. And so what these technologies provide to fab operators or improved yield and improved throughput at the leading edge.

So right now, that's where the battles are being fought at the leading edge, and we're winning every battle that we're engaged in. So we're in a very good spot. And so what these customers see the improvements we can bring at the leading edge, they want to see those same improvements at some of the other nodes they operate at.

And we've also seen migration from one device type to other device types. So we think this is a good thing for the company ultimately because it will allow us to basically ramp our new product revenue faster than we originally expected. We expect to see most of the new product revenue become meaningful, starting late this year, but really into '27 and '28 as some of these node transitions occur.

I appreciate that. That's helpful. And then second question from us. I actually wanted to focus on industrial and medical, which looks like it could represent either another leg of growth or help extend the growth profile beyond the next, call it, 12 to 18 months.

So I guess, first, can you just speak to the split that you're seeing between market growth and share gains? And then maybe you could touch on the status of potential M&A, rehash what you're targeting and how that landscape has changed since you first began the hunt to expand there?

Sure. Yes. So I think the good news regarding Industrial & Medical is the market has recovered. So we just went through a painful 2-year inventory correction period after the COVID supply chain issues. So based on what we see today, as our increased bookings, increased backlog, we think we're in very good shape from a market standpoint.

And we're seeing both the industrial and the medical markets pick up. And within those markets, test and measurement, aerospace and defense, factory automation, robotics and anything related to AI are leading the way. We think that the market share gain that we expect over the next 1.5 years is coming from new products. We have a number of significant design wins in our target segments.

They're going to help us pull ahead of the competition essentially. So I think we're in very good shape in I&M. As far as M&A goes, and I've said many times that our primary focus is to increase our breadth in industrial medical. And so we think this is a pretty fragmented market where we can in addition to our organic efforts, grow inorganically. So this will be an objective for us.

And we think some of the valuation mismatch we've had in past years, those mismatches are starting to close and that will allow us to make an acquisition sometime in the not-too-distant future.

Our next question is from Matti Hosseini with SIG.

Yes, it was on mute. Steve, just a clarification. You talked about 2 capacity expansion and talked about incremental revenue. Can you just tell me what your revenue would be once you're done with these capacity expansion projects?

Yes. So in my prepared remarks, I referred to the investments we're currently making in the Philippines, Malaysia and Mexico. And then we expect to be at a run rate or capacity revenue run rate of over $2.5 billion.

So this is our potential essentially after the investments in the current factory number are done. Second remark I made was on Thailand. So we are going to bring up Thilan earlier than we expected based on the strength we're seeing in data center in semiconductor. And so we think once Thailand is built out, that adds more than $1 billion of revenue-generating capacity.

So in total, I think we'll be in a position to ship in excess of $3.5 billion with our current factory network plus Thailand.

Got you. And this capacity expansion projects, when would it materialize? Would that be in the next 1 or 2 quarters? Or would you need more time for the expansion? .

Yes. So for the current factory network, it's -- the expansion effort is underway, and we'll have that in place in the second half of this year. So that's right around the corner. For Thailand, we'll start the investments late this year. So we're pulling in some of the spending we had expected to make in 2017 into second half of 2016 so that we can start up production lines for data center as well as production lines for semiconductor.

We're going to start that effort with large customers in with the higher volume products.

And the 43% margin target, is that inclusive of this capacity expansion projects?

It is. Yes. We comprehend the Thailand expansion in that number. really, that number is going to be driven by increasing new product mix. So as we introduce new products, they typically carry better margins than the older products. And secondly, we think we get better from a manufacturing efficiency standpoint. .

Sure. I squeeze 1 more follow-up. That actually as a follow-up to this capacity expansion projects. And just strategically, why not focus on executing on these expansion projects, executing on expanding TAM, especially on the data center, why dilute your effort by having M&A on the site or in Panama?

Yes, I think it's possible to do both. I think we spent the last 3 years basically streamlining our manufacturing capacity. If you remember, we broke ground on Thailand in 2023. So this has been the works for a while. I think at the same time, it's very difficult to reach our growth targets in I&M without acquisition.

So this is why we think in I & M in particular, it makes sense to push forward organically and inorganically.

Our next question is from Krish Sankar with TD Capital.

Steve, first question is on the data centers. I understand there's like quarterly volatility in data center revenue. Does we take your first quarter to analyze it, it's almost like low 30% growth. You're guiding to mid 30% plus. I'm just wondering, is the data center growth this year inhibitor due to the fact that components are constrained? Or is there something else going on in terms of the build-outs or market share or something else?

Yes, I think that's a great question. And as we look forward into 2026, the forecast, the unconstrained forecasts are quite strong, but our customers are dealing with downstream constraints. And so that's really tempering expectations for 2026. However, we think that some of these constraints will be addressed and so that would allow us to increase our forecast.

So I'd say in data center, there's definitely a bias to the upside in our '26 forecast. But again, some of these constraints need to be knocked down and then we'll be able to take advantage of a better short-term SAM. I think you may have noticed that our inventory is going up.

And much of that increase is us preparing to take advantage of these upsides that appear as our customers knock down these short-term constraints. So we are ready to respond if these constraints are addressed.

And actually, there is some of that in the first quarter. We were able to outperform in data center in the first quarter largely because some of the constraints were addressed by our customers.

Got you. And then a similar question on the semiconductor side. When I look at your full year revenue guide let's say, mid-20% revenue growth due in to mid-30%. It looks like semis is probably under growing what your debt and etch customers are talking about for WSP in the high 20%. So I'm just curious like what's going on in the semiconductor side. Are you just being conservative? Or do you think that semi growth could be actually high 20s to you to .

Yes. I think if you look over time, we had a good year last year, it was our second largest revenue year in our history. And so we came into 2026 pretty hot. And as you look forward in 2026, Actually, if you compare the second half of 2016 to the second half of '25, our revenue in semiconductor will be up more than 30%. And so we're going to be going very hot into 2027.

And so we're we are very happy with our backlog in semiconductor. And that's one of the reasons we're going to be opening in Thailand faster than we expected.

Lots that makes sense, yes. .

Our next question is from Jim Rashidi with Needham & Company.

I just wanted to go back to the data center. Growth that the moderation you're anticipating in Q2. Was there any pull-in activity into Q1? And as you talk about growth accelerating again? Is that something we should see as early as Q3? Or do you -- the constraints to your customers they're facing, would that mean the growth pickup is more in Q4?

Yes, Jim, I wouldn't characterize the Q1 outperformance as a pull-in. I think it was us taking advantage of an opportunity that was created when our customers were able to address their supply constraints downstream.

And so I think that's an indicator of what's going to happen in future quarters in '26. Very difficult to predict, but we know the demand mix is quite dynamic. We're trying to to stay in a position where we can respond quickly to changes in demand from our customers.

So I think that, again, there's upside potential in data center this year. and we're positioning to take advantage of that.

Okay. When you talk about the factory priorities, limiting the I&M I'll put in Q1. Wondering if you could size that. And so maybe elaborate, were these resources shift into other verticals. Can you talk about that?

Yes. Yes. So we build the I&M product in the same factories as we build the data center product. And so we had a surge in demand from data center in Q1, particularly in the last month of the quarter. And so our factories pivoted to data center. And unfortunately, we underperformed in Industrial Medical.

So I'm giving my personal attention to make sure we catch up to that demand over the next 2 quarters. So I think we'll solve this problem. The good news is we know what to build. Our backlog is robust now. So we're not taking a guess. We know exactly what the customers need and we're going to build it.

Okay. Do you expect to catch up in Q2 there?

Yes, it'd be Q2 and into Q3, and I think we'll be caught up. .

And maybe 1 final question just for me is, are there semi device types in particular that you're gaining traction with the new products?

So you're talking about semiconductor processes and so forth, Jim?

Yes.

I think we've basically focused on the leading edge processes, both in memory and logic. So we've been working with our customers very closely for the past 3 years. And also, our customers are also involved in the equation. So a lot of iteration going on. And what it's leading to is the realization that these technologies provide real benefits at the leading edge.

And so that's why we're excited about the potential to grow share based on these new technologies, and we can grow share in both memory and in logic.

Are you willing to share any kind of revenue expectation for this year from the new products collectively?

Yes. We haven't really guided to that, Jim. We said we made our goals last year. We expected to increase that significantly this year. But based on the timing of when these nodes start to ramp, we said that it'd be more significant growth in 2027. .

Our next question is from Scott Graham with Seaport Research Partners.

I have a couple of follow-ups on data centers and maybe I'll just ask you at all at the same time. So your sales in the first quarter were above your thinking. And it looks like that's going to sort of stabilize kind of go a little bit the other way in the second quarter.

And I understand it's a customer thing. But I guess my question is, why wouldn't customers -- I mean, data center demand is so dynamic right now, as you guys know. And it seems like these same customers faced some challenges yet we're able to kind of, for a lack of better term, fix them in the first quarter. Why won't that happen in the second quarter?

Yes, it's possible, that happens. But I think we do tend to guide conservatively. We see what's in front of us. We take the forecast into account and we typically will not include the upside in our forecast. But again, we're trying to stay as flexible as we can so we can respond quickly to our customers' change in forecast. So I think if there is an upside, we can take advantage of it. .

And I also know that your guidance for the quarter and the year in data centers was not including new customers. So I'm wondering if you have a flat period in 2Q, does that enable some revenues harvested from new customers to backfill?

I don't think so. So the position we're in right now with the second wave customers is that we've completed a number of qualifications in the first quarter, and we're working on additional ones this quarter. And those second wave customers are now qualifying our factories to produce those products.

That's typically a 6- to 9-month process. So right now, we expect the second wave customers to begin contributing meaningfully on the revenue line in 2017. But there is some potential to pull some of that revenue into the fourth quarter of '26.

I just -- and then I guess my last question is on the gross margin. So I think we -- in our conference call last quarter, you talked about hey, we think we're going to hit a 40% gross margin this year, implying one of the quarters. So now it looks like like you're going to not only hit it in the first quarter, but now you can hit it again in the second quarter off of your guidance.

And I'm just wondering is now a 41% gross margin, a point higher doable for 1 of the next -- 1 of the 2 quarters in the second half?

Yes. This is Paul. Yes, I think you're right about that. Clearly, we are a little bit ahead in Q1. We think we can build on that going forward because we are seeing traction on some of our new product mix. We're making progress in some of the factory optimization and other things.

And in general, as we have higher volumes, we should get some benefit. So -- when we look forward, we see that we should be able to improve gross margins modestly each quarter, which could definitely mean that we could get to 41% by the end of the year.

Our next question is from Elizabeth Sun with Citi.

I guess my first question is on the 800-volt transition. Steve, you mentioned in your prepared remarks, you are working with multiple customers working on some solutions. So I was just wondering if you could double click into what kind of solutions you are working on and what kind of customers you're working with and the rent -- the timing of the ramp in 800, I think your competitor is talking about second half ramping actually. And also, what will be the potent per -- as demand when the transition happens.

Yes. Yes. Thanks, Elizabeth. So right now, we're sampling our solutions to key customers we have a number of different options. So we have developed some 800-volt to 50-volt modules that provide 4,000 watt power 60-watt output power and 8,000 watt output power.

So we have different options for our customers. What differentiates us is very high efficiency, around 98% efficiency, high-power density and high reliability. So these are very important to customers.

And we've been told that we're leading from a technology standpoint. So we've got a number of customers that are very interested in technology. We're sampling it. We expect the initial production revenue, most of that will start next year. I think you'll see some small amount of revenue this year, but really, it's starts in earnest in '27 and becomes more meaningful in '28.

We think it's also good news for us because we think it increases our dollar content per rack. And generally speaking, these types of changes in technology are good for AE because we have a lot of the knowledge already in-house. So we're, I think, very well positioned as some of these data centers transition to 8 intervals

Got it. And the follow-up on the second wave of customers in data center. First question is just a clarification. If the demand puts in into the later half of this year, would that be incremental to the 5-ish percentage data center growth?

And also like how how should we think of the size of the ramp or like -- and the size of the volume potential. And also, if it pus in, do you have enough capacity to support the ramp, if that puts in early?

Yes. Elizabeth, this is Paul. Yes, so we don't have any forecast for our second wave customers in our guidance for 2026. So any of that that we're able to pull in would be upside to the 35% growth. And I think getting started on winning some of the designs and getting started on the actual manufacturing line qualification on that I think is pretty exciting for us because we're seeing, in general, that pull in.

We haven't quantified what that is. These aren't going to be customers that are the same size as our kind of primary customers, but they could be meaningful, and there are several of them. So on balance, we see this as a pretty significant driver or supportive to growth in 2027 from data center as we essentially expand our penetration across a broader set of customers.

Our next question is from Quinn Fredrickson with Baird.

Yes, thank you for the question. Just as we think about the data center outlook that you've outlined for 2Q and the back half, how should we think about supply constraints playing out from here? Or are you anticipating those sequentially ease at all or stay the same in order to hit the guide that you've discussed?

Yes. Quinn, the supply constraints I discussed so far have been downstream supply constraints. So they're really constraints our customers are dealing with. That's been the limiting factor, I think, in our forecast. So far, I think we're doing a pretty good job of managing our supply chain.

That's not to say we won't run into issues. But so far, nothing has really stopped us from shipping product. So Yes. I'm hopeful that our customers can knock down some of the supply chain issues later this year and we can increase our forecast.

That's helpful. And then Second, just on the new products. Curious if there's been any move forward in the timing of your dielectric etch wins converting to revenue at all?

I know 1 of your larger customers is recently discussing a pull forward of NAND conversion spending? Is there a potential that could move up the timing of new equipment orders and your wins in dielectric etch at all?

Yes. So on dielectric etch, we haven't been too specific about wins, but I'm very confident that we'll be able to communicate wins later this year. I think that what we're doing probably has little connection to the NAND upgrade exercise has a lot to do with next-generation nodes in DRAM and logic. .

Our next question is from David Duley with Steelhead Securities.

I guess first off, as far as the improvement in gross margins in the March quarter, that was done with the semi revenue being down, so I'm kind of curious if you could just elaborate on why the gross margin was better in Q1 given the semi mix was down.

And I guess a second question I have is regarding your new growth rate for semi in the second half of accelerating, I think, by 30%. Do you think that's more driven by your products actually starting to ramp up or is it more driven by just 2-nanometer spending is broadening out definitely beyond TSMC as we've seen in the news lately with both Samsung and Intel kind of joining the party?

Yes. Good question. This is Dave. I'll take the first one. Yes, margins were up in Q1. And I think the thing that we saw that was pretty encouraging was that at the product level, the mix was better. We're seeing a little better traction on our new products kind of across the portfolio, which carry better margins.

And so even though as a proportion, data center was higher and semi was a little bit lower. At the product level, we saw improvement. I think that's encouraging because, as you know, we're counting on that being a driver of gross margins as we move towards our 43%. And frankly, we think that can carry on through the year, and we should get some benefit given what we see accelerated growth in in semiconductor right now.

We also saw a little better, what I'll call other cost of sales, just a variety of things were a little bit better, including some cost of quality, made a little bit of progress on some of our ramp costs and efficiency. And we had a very modest pickup from tariffs. On balance, we had a few things kind of go the right way, and that offset kind of more broadly our factory ramp costs and slightly higher material premiums.

On balance, we feel good about the progress we're seeing. We think that can carry forward, which is why we've essentially modeled up slightly our gross margin outlook for the year.

Maybe I'll make a few comments on the revenue increase. I think as you look at our forecast for second half and the growth that we're forecasting is primarily due to growth in our flagship product revenue. When I say flagship, I mean the older products that have been designed in for a while.

We're also being helped by wins in the system power space with semiconductor testers as well as wafer fab equipment. The new products become significant starting in Q4 this year, but they become a bigger factor in 2027.

Maybe you could comment a little bit on what your customers are saying about growth in 2027. And if the 2-nanometer kind of ramp is spilling over into 2027? I would think that your semi business would perhaps go faster in '27 minutes grows in '26?

Yes. Yes, our customers in semiconductor are very enthusiastic about 2027. And I think one of the key factors is -- at the leading edge, there's a lot of new clean room space coming online in '27. And so they'll have a place to put all this new equipment.

So again, if you combine that with new product revenue becoming more significant in '27 due to the node transitions, I think it's going to be a very good year for Advanced Energy in the semiconductor space.

Okay. Final thing for me, you mentioned it twice in your prepared remarks and the Q&A here as the systems business for, I think, you said test semiconductor testers and other wafer fab equipment. Can you just kind of help me understand how that's different than the boxes you provided for your big OEM customers at this point?

Yes. Yes. So let me just explain that. We divided broadly into 2 categories. Plasma power is basically RF for pulse DC power we provide that's injected into a plasma chamber, and that's been our traditional business.

Now when we talk about system power, we're talking about the power that the machine uses to operate. So this is -- think of it as between the wall and the machine, that's system power. And so that could also be relatively sophisticated because these machines are very complex. And we've made a push into this area over the last few years and they're starting to pay off for us.

How big a TAM do you think this is?

Dave, what we have said is that in aggregate also plasma power is over $1 billion market opportunity for us. that's what we disclosed in 2012 by analyst event, and that's what we have said.

Our next question is from Brian Chin with Stifel. .

Maybe 2 questions from us. First, on the data center, definitely hear that the quarterly revenue progression in D.C. is more owed to timing and variability. But just thinking on a full year on your perspective, Increasing power content per rack wasn't a major tailwind last year.

It should still be a tailwind, but would you say it might be less pronounced this year relative to last year?

I don't think so because we continue to develop solutions for increased power. It's just a continuous exercise for us. And so I would say this is 1 of our advantages because the basic challenge is to continue to increase the power density at the same time, don't sacrifice reliability or efficiency. .

So basically, you have the same amount of space, but you have to produce a lot more power. And to do that reliably and efficiently is a challenge from a design standpoint. And so that's one reason we've been able to be successful with our hyperscale customers, and we've been successful now with the second wave customers because we have that combination of technical features as well as good speed of development.

And then finally, we've got the factory network to produce these boxes in high volume.

Okay. And then second question on semi cap and answer this however you're comfortable addressing it, but in terms of that second half guide that you're providing. Do you think you're shipping in sync with customers or perhaps a little ahead given that multiyear visibility and outlook?

And then kind of second part of that is from a supply chain perspective, given the duration of visibility certainly well into next year, has that given you a lot of confidence to go out and maybe be really proactive sourcing some of those components, maybe call it chips, other materials that can be issues as you get through like a multi-quarter growth period?

Okay, Brian. So on the customer demand being in sync with their requirements or are they putting in inventory essentially. Our view is that we're more or less in sync. I think what we've seen is across the customer set that we've received increased orders. And that's because of leading the edge primarily.

So we know the ENC market demand is there. I think the constraint right now in semiconductor is really clean room space. And so if some of these clean rooms come on faster than expected, then that will increase demand and if they come on slower than expected, then we may create a little bit of inventory, but I don't think it will be much.

Now as far as supply chain goes, it's a hot topic, right, especially after COVID when we had to deal with so many issues. So we have been leaning into inventory so that we are able to have some reserve in case things get tighter, lead times go out.

So I think we spent quite a bit of time on the supply chain topic. And one of the things we did during the COVID supply chain shortages was to develop more second sources, which we've done. So that gives us more flexibility and where we couldn't develop a second source, then we put in extra inventory.

So I think we're going into this with our eyes wide open, and we're pretty aggressive on the inventory front.

Our next question is from Jim Rashidi with Needham and Company.

Paul, you may have given this, I apologize if you did. But any color on OpEx as we look out over the balance of the year beyond Q2? Anything we should be thinking about?

Yes, Jim, obviously, OpEx was down a little bit in the first quarter. So we've done a good job controlling spending there. But in the second quarter, we do have our annual salary or merit increases.

So we've guided second quarter of $5 million to $7 million. We continue to expect that, that run rate will increase a little bit every quarter, kind of what we said before, as we invest in these new programs, primarily in engineering as well as some variable costs with higher volumes.

So we would expect operating expenses to be in the $460 million range for the year and kind of graduating up kind of sequentially to basically at that level.

Thank you. This concludes our question-and-answer session and our conference for today. We thank you again for your participation. You may disconnect your lines at this time.

Hi, everyone. I'm Shane Brett, U.S. semiconductor equipment analyst. Joining me today from Advanced Energy are Paul Oldham, EVP and CFO; Edwin Mok, Senior VP, Strategic Marketing and Investor Relations.

I have a quick disclosure read. For important disclosures, please see the Morgan Stanley research disclosure website at www.morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representative.

Before I start my question, I hand it over to you guys for disclosure.

Sure. So just quickly, obviously, we just reported our results on February 10. So obviously, any outlook that we provide is based on what we reported. There's no update to our guidance. For any risks and uncertainty factors, feel free to check our SEC filings.

Great. So I want to start with some longer-term questions. So you hosted your Analyst Day about 1.5 years ago where you outlined growth across end markets past the $3 billion in revenue by 2030. Just a lot has changed in that 1.5 years. So can you discuss how the industry environment has evolved, how that's sort of influenced your sort of forward outlook?

Yes. Great question. Obviously, it was an exciting year last year, and we think it's going to be an exciting year this coming year. But some things have changed. Data center has grown much faster than we expected. We more than doubled in that market last year based on new products that we brought out, our new focused strategy on really looking at areas within that market where we could be more differentiated and a good embrace of our products.

Semiconductor also grew last year as a second year of growth for us, and we expect it to grow again this year. And of course, I think everybody is getting more and more excited about the opportunities in semiconductor and WFE, particularly in the second half of this year and into next year. Industrial & Medical is probably the one area that's a little bit behind. It had a pretty tough start last year. We bottomed out in the first quarter. But since then, we've seen sequential growth every quarter. And in the fourth quarter, we saw a 10% sequential growth, and it was the first quarter of year-over-year growth in 2 years.

So that market is largely normalized, and we think it will grow pretty steadily from here, kind of paced by the health of the overall macro economy. So on balance, we think we're a little ahead. We grew 21% last year. We guided to high teens growth this year. And there's potential for upside to that depending on kind of how the market goes. over the next few years. So on balance, we feel pretty good about where we're at from a revenue perspective.

I feel like saying a little ahead might be underselling yourself as your 2030 model implies $2.5 billion revenue, and it's probably not inconceivable that you could reach that level as early as 2027. Just relative to the financial model you outlined for 2030, in a world where that $2.5 billion is earlier, just what could gross margin and operating margin look like, just given there may be some longer-term initiatives there?

Yes. Great question. Gross margin, we've made a lot of progress there. We moved from about 35% margins at the beginning of '24 to exiting 2025 at almost 40%. So we think we're well on track to our goal there. Obviously, tariffs are new. That wasn't contemplated in our model. That's 100 basis points roughly impact. And frankly, the data center mix is also growing, and that's good margins for us as we changed our strategy, but of the 3 markets, that's the one that's lower and it's approaching corporate average. At the same time, we've made really good progress in our manufacturing consolidation and other actions that we've taken.

The biggest lever for us going forward on gross margin is going to be our new products starting to take more traction in the market. We think we can get 200 to 300 basis points more benefit from the portfolio mix there. And then finally, volumes. I mean, obviously, we're getting the benefit of higher volumes, and we'll continue to see that from this point forward. I think on operating expenses, while we might be made good progress on gross margin, we have a couple of headwinds there. We may have a little bit of tailwind in operating expense if things are growing this fast. We said we'd grow revenue or operating expense at half the rate of revenues.

Last year, we grew revenues 21%, and we only grew operating expenses about 7%. So we're probably a little ahead of that game if we can get there earlier because we won't have 5 years of growing expenses to support that revenue. And we don't need to make any new fundamental large investments to get to that $2.5 billion. Our products seem to be doing very well. I think below the line, there are some pluses and minuses. The biggest one, of course, is our interest income or other income probably doesn't quite get to our model that assumed 5 years of growing cash and getting interest, we might have less. And then based on the performance of our stock, we have some dilution to the to the share count. So I think those are kind of the pluses and minuses. But on balance, we're very excited about where we are and our ability to grow earnings at a faster rate than was in that original plan.

And let me drill into that kind of the investment needed for growth and how that looks like for CapEx. As you've sort of indicated that 2026 CapEx will remain at the Q4 run rate and you're able to supply $2.5 billion of revenue capacity. But just what's the CapEx involved in sort of supporting this kind of dynamic growth environment right now?

Yes, it's really 2 things. One is capacity growth. Obviously, we're getting, particularly in data center to the numbers much faster than we had originally anticipated. So think of that as a pull-in of capital we had already planned into more like late '25, '26 and maybe early '27 versus more spread out over the 5 or 6 years.

Second is investment in capability. With the explosion in kind of the power levels that are required for our products, we're investing in capability of just more power in our factories, more power in our R&D centers. So it's those 2 things that are the primary driver. We'd already contemplated the expansion in our number of sites. So the Philippines was always contemplated in our [ 2030 ] growth numbers. Obviously, with the growth, we'll probably be opening that earlier than we probably otherwise would have. But we expect after this next maybe 4 to 6 quarters of higher spending, we would see that moderate back down. Again, a lot of this is a pull forward based on the fact that we're seeing.

The Thailand capacity coming online. Is that sort of -- how should I think about the Thailand facility relative to the kind of $2.5 billion you guys are prepared for right now?

So the Thailand facility would be 100% incremental to the $2.5 billion in revenue. The status of that is we did start that investment back in 2023. So we kind of thought ahead about what we would need. So the factory is built out. It's basically finished. What it needs to be done now is just facilitized for manufacturing. We can do that in about a 6-month process. So from kind of a go signal, we can have lines in that factory producing product. So we'll look at that, the timing of that a little bit.

We've said right now that we would largely use that for the second wave customers in data center, which would be qualifying this year and would look to ramp next year. So it's likely an early '27 opening of the Thailand factory. And we could accelerate that or slow it down based on how we see the market evolving. I'll also note that, that's a good place for us to add semiconductor capacity. When we first started that facility, we actually thought semiconductor would be the first products in there. But as you know, this very exciting ramp we're seeing now had been predicted for each of the last 2 years as well and didn't quite materialize.

So we'll probably start with data center products there, but be looking closely at semiconductor as kind of a business continuity additional capacity there as well.

Okay. I know you guys are a data center stock right now, but I'll start with semi equipment and on [indiscernible] markets, but...

We're happy to be both.

Okay. So I've hosted your biggest customers just over the last 2 days, and they're very excited over the next 2 years. I had one of your customers say, "Hey, we have 8 quarters of visibility." And another customer say, "We have a lot more visibility than we did this time last year. It just looks really good."

Just at your recent earnings call, you mentioned that Q4 -- so the December quarter revenue upside was not from pull-ins from '26. You expect Q1 '26 to grow quarter-over-quarter and that second half '26 will be stronger than first half '26, almost just a sequential increase through the year. Just can you walk us through what are those customers that have been on stage the last 2 days telling you? And what are you prepared to do in '26?

Maybe I'll answer that at a high level, and then Edwin could talk a little bit about some of the details. But you're right. I mean, even as late as our earnings call in -- for Q4, which was in kind of early November, we talked about the semiconductor market declining a little bit. Since that time, obviously, we ended up growing 8%, so a new level. And we said Q1, where we thought it would stay at the lower level is actually going to -- most of our Q1 growth will come from semiconductor. So definitely, the tide -- the water level has increased. It's not like a pull-in and a onetime thing. We see that the water level is increasing, and it seems to continue to increase.

Obviously, I think the comments from our customers, even some of our peers have been pretty positive. I think that gives us increasing confidence that the second half will be higher than the first half and that it's a good setup for another strong growth year in 2027.

Yes. In terms of growth in WFE, I mean, typically, we don't predict WFE, right? As a company, that's kind of our policy. But generally, there's a pretty big consensus that generally, you are seeing strong pull for both leading edge foundry logic as well as in the DRAM side. On the DRAM side, historically, the last few years, there's a lot of investment in HBM, but it seems like this year, there's incremental spending on wafer, either wafer start or moving to advanced node process.

The good news is that these leading-edge process, we generally have higher dollar content. So that should be positive for us. The flip side is we do have lower content for NAND, which seems like it still grow this year, but it's mostly on upgrade, which we generally have lower content. And also trailing edge, we do have healthy exposure. I think the consensus is around flattish this year. So in aggregate, I think some of the growth trend this year is positive for us for where we sit. So we should benefit from that. And in terms of beyond this year for '27 and beyond, I think we -- I read all the reports that Shane writes and how foolish it is. So I'm all for that.

Okay. I appreciate that. Those are my numbers, not yours. So I want to hold you guys accountable to those.

So I guess one thing I would just add to that, though, Shane, I think it's important is, obviously, this industry is inherently hard to project a little bit because you do have the sort of everybody moves in mass. But the one thing that we've tried to do short of being prognosticators of what the actual numbers are is to be prepared. So we mentioned in our last call that we're leaning into inventory, both for semiconductor and data center so that we can capture upside of business that's coming. We've just talked about capacity. We made capacity investments. We've retained people through the softer parts of the industry. And so we're well prepared.

So if WFE grows faster, that's great. We'll be prepared to participate in that. And all the trends that you've talked about from a technology perspective, from a leading edge node perspective, those directly benefit us as well. So we're pretty excited about where things are going, and we'll be prepared.

So I guess across the semiconductor ecosystem because of AI, there's always a bottleneck somewhere you guys are not going to be that bottleneck because you're prepared.

We are not going to be the bottleneck. That's our goal.

Okay. Great. I guess kind of when I think about your growth relative to the customers, a dynamic that I have to consider for subsystems is inventory. And throughout the course of 2025, some of your customers did lower inventory, but now that we're sort of headed into these 2 strong years of growth, there's -- it's probably unlikely that they kind of lower inventory. What are they kind of telling you on how much AE products that these guys need right now?

I think on balance, what we've seen is the -- as we talked about in our call, the forecasts have increased. And I think they've been public about talking about having their suppliers get ready. So we hear those messages as well. We've said recently that inventory levels are largely normalized at our customers. And so we think we're sort of feeding at the demand level. Now having said that, it's typical that we would see higher volumes a little bit ahead of them, maybe a quarter or a part of a quarter because as their volumes ramp up, they need to make sure there's enough on the shelves for them to pull.

So we think we're heading into this from a normalized inventory position. And as we see this growth accelerate, we should be able to keep pace with that. If they start to add additional inventory in front of that, then we could see a little bit of the benefit of that early.

And when I also think about your growth this year, kind of a good aspect of that is also just new products. Just where are these sort of new products? Can you kind of give the audience a little bit of clarity on where the new products are kind of generating these wins? And just how are they going to layer on to the growth this year?

Yes, Edwin, why don't you jump into that?

So in terms of new products, we launched these new products, eVoS, eVerest and [indiscernible] in 2023. These are next-generation plasma power generator that we believe is substantially advanced versus current generation of product versus our as well as our competitors. And therefore, we've been working very hard to secure design win. Unfortunately, in this space, it takes a few years to win the design to translate to revenue. The good news is that we start to have some win in 2 areas that we've historically had stronger position, conductor etch and deposition. And those wins start to translate to initial revenue in 2025, which we expect to contribute greater revenue in '26 and beyond.

These initial wins are largely at 2-nanometer as well as kind of [ 1z DRAM ] node, which is just a limited number of steps. Typically, that's how it works in the space. The first win is going to be of nodes. But our customers have our product design into their tools, and they are working on winning the next node, which is sub 2-nanometer and also 1D and beyond. And we believe there will be more applications for those nodes as they ramp. So overall, for what we have won and what we expect to come just from conductor etch deposition, we expect incrementally higher demand as we go through the next few years. It depends -- partially depends on timing of when those next nodes will come.

In addition, in dielectric etch, where we have very low, almost no share in dielectric etch, we have been working hard with multiple customers on multiple programs to win those. Initial feedback or feedback from our customers has been really positive, and we believe that it really demonstrates the yield and throughput benefit of our product that we can bring to bear at the wafer fab. We're still working towards that. There's nothing to report at this point, but we expect some wins which should translate to initial revenue as early as '27 and beyond.

So I guess to kind of sum up your answer, the areas where you have had historical success in conductor etch and deposition, you're sort of strengthening your share there and sort of seeing if there's other end markets that you can kind of penetrate. And then -- so you've also mentioned system power ramping this year on your latest earnings call. Just can you give us some clarity on just what are the applications for those wins? What's the kind of competitive landscape for those products?

Okay. So system power is actually power solution that we offer that power other part of the equipment. It could be powering the actual equipment itself, right? It could be power heater or it could be other applications such as test and measurement or ATE or assembly and semiconductor equipment space. We historically are not in that space. But through a number of acquisitions, including Artesyn Embedded Technology that we bought in 2019, we have the product that actually go after those opportunities.

Over the last few years, we've increased our focus into winning some of these opportunities in some of these applications. And we won some design that we expect to start generating revenue as early as this year. On our analyst event in 2024, we have set a goal for this portfolio of product to add an incremental $40 million of revenue by 2030, and we believe we are well on the way to achieve that.

So this will be revenue with the semiconductor test players in the back end?

It's both our leading wafer fab equipment customer, where it's used not for generating plasma, but it's to power the system, different part of the system or could be heater or other pieces or it could be like the test manufacturer.

Got it. Got it. Before moving on to data center, just -- as you think about where [ AEIS ] is allocating R&D dollars within semiconductors, which areas are you giving the most focus? And can you kind of give us a glimpse into sort of the road maps that your customers are outlining for you guys?

Yes. I'll maybe take the first part and Edwin can talk about the road maps. But clearly, our main investment is in our core area of plasma generation. I mean this is -- it's the largest part of the market. Frankly, it's very hard to do, particularly you get into these next-generation nodes. These are quite complex products. And they take a lot to win. But as you know, once you've won them, it's a decade-long payoff or longer. And so we think we're -- I mean, Steve said in our last call, we think we're better positioned as a company than we've ever been in semiconductor, not because we're generating a lot of revenue from these new products right now, but because we see how the qualifications are going and how that sets us up for success for many, many years to come.

Yes. In terms of road map, I think we launched this product in '23, partially because we start to see increased demand for more complex etch and deposition process. A lot of these devices are moving to 3D device, which then requires much more -- number one, more number of steps. And second thing is higher power requirement, higher plasma requirement. In one example, for example, much higher and much narrower high aspect ratio etch process, actually, both conductor etch and dielectric etch that we're seeing.

So we believe our product has allowed our customer to achieve some of those goals. And again, beyond the process itself, also allow them to maintain the throughput because as you etch deeper, it slow down your etch process and also allow them to achieve the low level and therefore, the high yield level that they could achieve in the existing technology. So I think this transition to 3D really plays well and really result in our customers pulling these new products from us. And I think we think we're well positioned to on that.

Got it. Got it. Let's move on to data center. So you're guiding more than 30% growth in 2026, but that's after doubling in 2025. Just -- but we're also seeing just hyperscaler CapEx forecast that might be a little bit higher than 30%. Just can you discuss the gap between this hyperscaler CapEx growth and your own data center revenue growth? Just give us a little bit more clarity around that.

Yes. It's a really good question. We get it quite a lot because I don't think -- it's difficult to draw exactly what that correlation is. Obviously, in 2025, we won a lot of new designs. I think we really established ourselves with our strategy of focusing on some of these higher precision, higher differentiated applications. And we grew over 100%, which was much faster maybe than the CapEx growth.

A lot of those designs are in place, and we're winning the next generation of those in 2026. So you don't see quite as fast of growth related to that per se. But we're very excited about our position, and we think there's opportunity potentially biased to the upside based on kind of some of the demand signals that we could be seeing. So the last thing I would say is that there's a lot in the hyperscale capital projections, right? They're investing in buildings. They're investing in GPUs, which are going up in price significantly year-on-year. Memory, we know prices are doubled.

So by the time you get down to the rack and to the power in the rack, we're a pretty small percentage of that total number. And remember, we're focused on the more precision areas. So I think our growth will be more driven by our ability to continue to expand the places that we play and get those next-generation wins than necessarily just aggregate growth. Obviously, that's a good sign. That means they're putting -- continue to put a lot more money into these. But I think we have to think of ourselves a little bit more like based on our targeted approach to these markets.

Got it. And kind of when I think about this year, if I'm correct, you have 9 months visibility right now, which probably takes us to the later point of this year. But what are these customers, I guess, telling you about '27, '28? Just can you provide any color on just what those discussions with those customers might be looking like?

Sure. So in terms of like demand visibility, we have our customers' demand planning for this year. That's what we use as an input to come up with this up 30% growth, right? Our customer is telling us that there's a lot of volatility in the supply chain. In fact, we comment about this for the last 2, 3 quarters in terms of that result in the demand or the type of product they demand from us changing pretty rapidly sometimes. And we have to flex our manufacturing to meet that demand. And that is also factored in this up 30% guidance that we have.

In terms of kind of outlook beyond that, I mean, obviously, the AI trend is pretty strong, right? And then our customers want to make sure that we are prepared for -- ready for long-term growth. That's part of the reason why we are building additional capacity this year so that we have this $2.5 billion capacity to meet that high demand, with Thailand be ready for additional ones, right? So overall, the trend is positive. And last thing to point out is in terms of technology development, we're also working on technology that is ready for '27 and '28 beyond technology launch. Some of the programs that we are ramping this year are stuff that we won last year, and we have been going on this annual cadence of winning design that expect to ramp for the coming year. So we expect the same kind of cadence for probably this year and next year at least.

Yes. I think the aspect of continued technology evolution in terms of power requirements, architecture, all of those things are as important as the absolute investment that's been making because that's driving change in the industry, which is going to drive continued investment to get to the next generation and that plays in our favor because that's what we do. That's what we do as a leading-edge technology provider is stay with our customers as they make these rapid changes. So we think as long as that pace continues as well, that's going to continue to drive growth for us in the industry.

I guess I want to talk about that point of just, I guess, you're seeing accelerated growth, but it's mainly with kind of a select few customers. You've talked about this kind of new second wave of customers coming that might not be as sort of engineering intensive as you can apply kind of your learnings from your first group of customers to your second group of customers.

I want to hear about that kind of how that sort of, I guess, investment for these customers might change with the second wave of customers. And the reason why I'm asking this is data center has been a higher growth segment, but a lower gross margin segment. Just how does that change with the second wave of customers?

And maybe starting with the last part of your question first. I think from a gross margin perspective, we would assume it's the same or better. That's the first thing. But I think on a total contribution perspective, it should be substantially better because essentially, what we're able to do is take technology blocks at the very leading edge or these very specific applications and pivot them relatively easy to another set of customers who are now looking at that and saying, "Hey, that fits really well with the requirement, I would like." We've already developed it. And so we're able to pivot that to them with relatively few changes.

So there may be a little bit of customization or tailoring for exactly their former fit, but we're essentially able to take leading-edge products and/or technology building blocks and repurpose them to these other customers who are coming along in the wake of the investments that the large hyperscalers are making. So there's a lot of R&D leverage from that perspective, which is important for us because we want R&D engineers focused on the leading edge. And then we want to repurpose that into this next set of customers that are coming along. And we'll still have a very differentiated product, performs extremely well, but we're able to leverage that investment that we've already made.

So it's a quite margin accretive development that you're hopefully going to see over the next few years with data center.

Certainly, from a total margin, that's correct. And from a gross margin, it should be the same or better.

Okay. Got it. So it's just kind of R&D is probably less for the second wave.

R&D will be substantially less.

Okay. Got it. Got it. I want to switch over to Industrial & Medical, and then I'll open it up to questions. But 2025, a bit of a digestion year. You're forecasting sort of revenue growth in '26. Just what's driving this recovery? Is it inventory normalization, new product adoption? Can you kind of give us some color there?

You want to take that one, Edwin?

Sure. So 2025, Q1 was the bottom for the industrial medical revenue. Since then, we have 3 quarters of sequential growth. And throughout that period, we track, especially our distribution data to see kind of how healthy the industry is, if you will, right? And we've seen both -- over the last few quarters, we have seen channel inventories continue to come down. We've seen our bookings start to pick up, especially in the latest quarter as well as our order book start to expand.

So we think that we are at a point where channel inventory is largely normalized demand, both our sell-in as well as sell-through start to approach a normalized level as we come to this year. That give us confidence that we can have sequential growth over the next few quarters. In terms of demand, with the exception of the -- obviously, we are seeing demand for existing products that we have already sold in. On top of that, last few years, we've been working on some new design. And some of them, we believe we will start to see a ramp. On the recent call, we highlight, for example, some design that we won in defense and aerospace as well as some additional design that we have won in the factory automation space.

Some of the design we won them because the industry is going through or the channel is going through inventory digestion, customer doesn't want to buy the new stuff, right? But now that we are normalizing, we are starting to see this new design ramp. So that combination give us confidence that we can see growth this year and probably extend beyond.

I just want to clarify on that. So for the sequential growth outlook, is that kind of your customers giving you strong orders? Is that kind of lead times extending out? Like what kind of gives you confidence on that sequential growth outlook?

Yes. I would say it's a combination of things. One, we are seeing more orders. So we reported much higher orders in the fourth quarter. Those weren't all for deliveries in Q1. We are seeing customers look out on the horizon now and say, okay, I want these products over this period of time. So I think the order book has clearly strengthened. With that, the backlog strengthened.

I also think when we look at the other metrics about inventory at distributors, as Edwin was saying, sort of a sell-in versus sell-through, those things are largely normalized today. And so as this market grows, even if it's modestly, we should participate in that at that level. And then there are some parts of the market that are growing faster than others. So I think all those factors contribute to seeing kind of continued steady growth. If the market gets better, that's great. Edwin mentioned our new products. We're actually seeing some uptake of those new products already. It's just overshadowed by the depth of this kind of digestion period as well as some winnowing that we've done within the portfolio.

So as the market recovers, we should actually start to see a more disproportionate impact of the new products helping to drive growth.

The audience has any questions. Sharon, there's a mic right behind you.

I just have one question related to your power offering for data center. We all know that the power architecture for data center is kind of moving toward high-voltage DC. What does that mean for your offering? And would you -- when would you expect the contribution will start to come...

Yes, I'll take that question. So on the last 2 earnings call, we actually talked about working with our customers to develop next-generation power solution. Typically, frankly, this development takes 2, 3 years. So we are already working on things that we expect to ramp in '27 and '28, right, because you can't -- otherwise, you can get those out by '27, '28 realistically. And some of those applications or actually a good chunk of that is related to HVDC. In terms of what impact it has, we will have an advanced how we kind of think about it. We generally believe that high-voltage DC will translate -- number one, it increase the power level per rack generally higher power level per rack means we have a higher dollar content per rack, right? So that's positive for us.

In addition, we believe we are developing some unique technology and based on initial feedback from our hyperscale customer, lead us to believe that, that might allow us to further expand our footprint across our hyperscale customer and eventually other customers. So there's a potential SAM expansion for us as a result of adoption of high-voltage DC. As Paul described, our first goal usually is to work with the hyperscaler to partner with them to develop the technology. That's still going to be the case. And based on the program that we're working with them right now, we don't expect production revenue to come until maybe second half of '27 and more meaningful in '28 and maybe '29 for different variants. There's a few different flavor approach that people are taking. Naturally, eventually, some of those might not fully ramp, right? So -- but we are working on multiple programs at the same time.

I think the exciting thing about HDC is it is more power per rack. And what's required are exactly the things that we're good at, high power efficiency, high power density, the ability to work with an engineering team to get product to market and then ramp that quickly as needed. So we see that as an opportunity. That should create more opportunities for us that play to our strengths.

So it's a very kind of attractive content per rack opportunity for you guys?

We believe so. I mean, obviously, as Edwin said, that architecture is still being very much defined and what it will look like is being defined. But based on the early discussions that we've had and the development work we're doing, we think that's right.

Got it. Got it. Any other questions? I guess then I'll ask a financial question. So you've historically pursued opportunistic M&A, but it's focused on Industrial & Medical markets, but the kind of explosive growth that you're seeing is for data center, then semi is probably a second. Just has your capital allocation or M&A priority shifted given kind of these dynamic end market shifts?

Yes. If you take a look at total capital allocation, we've said the majority of our capital we want to put towards growth. The reality is for the data center and semiconductor markets, that's largely internal investment, and you see a lot in the form of CapEx. So we are, I'll say, doubling down on those markets. I'll say, from an internal organic perspective. There is also some technology things we can pick up there. Like we bought Airity Technologies, I guess, almost 2 years ago. 2 years ago. That's highly embedded in some of our next-generation products, and we're using it across the portfolio.

So there's opportunities certainly in those markets. But if you think about large-scale M&A, the semiconductor and data center are very concentrated. And while there's things along the periphery, they're not really our -- what we're good at. And so that kind of leads us to Industrial & Medical. Industrial & Medical is almost the opposite. It's highly fragmented. That market has all the characteristics that we like being like designed in, long life cycle, sole sourced, good margin. But because it's so fragmented, the organic growth on that path is probably slower. You're not going to have these periods of explosive growth. So the way we think to grow that market is building a bigger beachhead, and you can do that through M&A. And in this case, we're able to basically take a broader product portfolio, better customer access, more scope, if you will, and leverage the scale we have as a company.

So the synergies in this market would be very natural where we can essentially expand our offering without and being able to rationalize some of the infrastructure and the scale. We think it's important to have 3 strong pillars. Obviously, data center has been all the rage, semiconductors becoming the rage. But I'll remind you that in 2023, we performed much better than our peers because Industrial & Medical actually outperformed both those markets substantially. So over the long run, we think it's a good ballast for the company to have a strong position in all 3 markets. And so we're very committed to that M&A inside Industrial & Medical. We're also prudent.

So I think the valuations have been a challenge. And so we've taken our time to make sure we look at the right opportunities, and we create value with those. And we've said recently that as the I&M market has normalized, that maybe is creating a better baseline to have more constructive discussions from an economic perspective. So we're -- it's an area we're committed to, and we'll continue to invest in, but we'll do it wisely.

Got it. We have a minute left. But -- so you've sort of transformed into kind of you're in a great neighborhood and you're building a great house in this great neighborhood of data center and semiconductor equipment. But is there anything that you think the financial community might be sort of underappreciating with Advanced Energy in any way?

Yes. I think this idea that we're becoming a more diversified company that has more paths to growth is actually starting to sink in. And we're seeing that obviously in revenue growth. We're seeing that in earnings growth. And so I think that's probably the most important thing is we have multiple vectors for growth. And particularly as we look at the next year or 2, we think we're in a period of time where all 3 of these markets could grow.

Got it. Great. That brings us to time. Thank you, Paul. Thank you, Edwin. Appreciate it.

Thank you, Shane, and thank you, everybody, for joining us.

Appreciate it.

Greetings, and welcome to the Advanced Energy Fourth Quarter 2025 Earnings Call. [Operator Instructions] As a reminder, this conference is being recorded. It is now my pleasure to introduce Edwin Mok, Senior Vice President of Strategic Marketing and Investor Relations. Please go ahead.

Thank you, operator. Good afternoon, everyone. Welcome to the Advanced Energy Fourth Quarter 2025 Earnings Conference Call. With me today are Steve Kelley, our President and CEO; and Paul Oldham, our Executive Vice President and CFO. You can find today's press release and earnings presentation on our website at ir.advancedenergy.com.

Before we begin, let me remind you that today's call contains forward-looking statements. They are subject to risks and uncertainties that could cause actual results to differ materially and are not guarantee of future performance. Information concerning these risks can be found in our SEC filings. All forward-looking statements are based on management's estimates as of today, February 10, 2026, and the company assumes no obligation to update them. Any targets beyond the current quarter presented today should not be interpreted as guidance.

On today's call, our financial results are presented on a non-GAAP financial basis unless otherwise specified. Detailed reconciliation between our GAAP and non-GAAP results can be found in today's press release.

With that, let me pass the call to our President and CEO, Steve Kelly.

Thanks, Edwin. Good afternoon, everyone, and thanks for joining the call. We finished a very successful 2025 with a strong fourth quarter. Revenue of nearly $490 million was at the high end of our guidance. Strengthening demand in the semiconductor, industrial and medical markets drove the outperformance.

As expected, we also had another record quarter in data center. Gross margin came in just shy of 40%, our best performance in 5 years. Earnings per share of nearly $2 also beat guidance.

For 2025, we grew total revenue over 20%, increased earnings per share by over 70% and significantly improved our gross and operating margins. We also delivered record operating cash flow.

Our strong financial performance underscores the benefits of our diversification strategy, our focus on execution and the leverage in our model. We deploy our best-in-class technologies across multiple high-value markets, allowing us to deliver healthy revenue, profitability and cash flow through market cycles.

In 2025, we grew revenue in two of our three target markets. Data center computing revenue more than doubled year-on-year and increased sequentially in every quarter of 2025. Hyperscalers have adopted our customized power solutions in a variety of AI rack applications.

Semiconductor revenue grew 6% year-on-year to the second highest level in company history. New products began contributing incremental revenue in 2025, as some of our design wins moved into early production.

Although Industrial Medical revenue declined year-on-year, we were encouraged by [ 3/4 ] of sequential revenue growth after reaching a bottom in the first quarter. We expect growth to continue in 2026 as many customers and distributors have worked through excess inventories. We also think that our design win pipeline will drive share gain moving forward.

In 2025, we maintained a solid cadence of new product introductions with 26 new product launches across our markets. In addition, we spun off many custom products.

In semiconductor, we continue to receive very positive feedback on the best-in-class performance of our eVerest, eVoS and NavX technologies. At the leading edge, these technologies are delivering meaningful improvements in yield and throughput. In addition to our many confirmed design wins, there are a number of development projects currently underway, which should convert to wins in 2026.

In addition to our success in plasma power, we have also made progress winning key system power slots for semiconductor equipment with multiple wins ramping to volume this year.

In data center, our 2025 wins are going into volume production this year. Working closely with key customers, we are developing new technologies and products for next-generation AI data centers. We are also engaging with a second wave of cloud and enterprise customers, largely with modified versions of our standard technology platforms.

Through the prolonged inventory correction in the industrial medical market, we continue to invest in new products, customization capabilities, digital marketing and distributor partnerships. Now that the market is recovering, we could leverage those investments to gain share.

In operations, we expanded capacity in the Philippines and in Mexico, enabling us to support the continued growth in data center demand. We also completed the fit-up of our new Thailand factory, which is expected to deliver more than $1 billion in annual revenue-generating capacity once it's fully built out.

Through solid execution, including the closure of our last China factory, we expanded gross margin by 240 basis points. Despite ongoing tariff headwinds, we are well positioned to move gross margin above 40% in 2026.

Now let me provide some fourth quarter commentary. In semiconductor, fourth quarter revenue grew sequentially, well ahead of plan. In data center computing, fourth quarter revenue increased sequentially to a new record, driven by AI data center investment.

And in Industrial and Medical, revenue grew 10% sequentially and returned to year-over-year growth after multiple quarters of decline.

Bookings, backlog and resales were up. Channel inventory was down. We believe that the market environment for Advanced Energy has largely normalized.

In Q4, we secured important design wins in factory automation, medical imaging and electrosurgery applications.

Now I'd like to provide our view on 2026. Entering the year, we see positive demand trends across all of our target markets. In addition, we expect that multiple new wins will ramp to production in 2026, driving growth across the portfolio.

In semiconductor, stronger customer forecasts are increasing our confidence in a strong second half. In addition, we expect new product revenue to grow over the course of 2026. These forecasts are underpinned by downstream investments in advanced logic and memory capacity.

In Data Center, we now project full year revenue to grow more than 30%. Modular technology blocks, strong design team and development speed are key enablers of growth in this market.

In the industrial and medical market, we expect demand to continue to improve over the next few quarters. Production revenue from several wins in factory automation and defense should enable us to outgrow the market. In total, we project that our 2026 revenue will grow in the high teens after 21% growth in 2025.

Let me finish with some closing thoughts. Advanced Energy designs and manufactures precision power solutions for demanding high-value applications across multiple markets. Our market diversification strategy, coupled with aggressive investment is enabling us to capture upside across our markets and deliver more consistent financial results. We've steadily increased our R&D and marketing spending over the last few years, building a strong portfolio of new products, gaining new customers and growing our design win pipeline.

We have more than doubled the output of our Philippines and Mexico factories. In addition, we've built a new flagship factory in Thailand.

Finally, with a strong balance sheet, we will continue to pursue inorganic growth to improve scale and to broaden our technology portfolio. With market tailwinds in 2026, strong demand for our new products, expanding margins and a solid balance sheet, we are confident that we can meet or exceed the long-term financial goals presented at our 2024 Analyst Day. Paul will now provide detailed financial information.

Thank you, Steve, and good afternoon, everyone. 2025 was very successful for Advanced Energy. We delivered double-digit year-over-year growth for both revenue and earnings throughout the year, led by record data center revenue in every quarter. We executed on our gross margin improvement plan, exiting the year approaching our initial target of 40% despite the impact of tariffs. Disciplined spending helped drive operating margin to its highest level since 2022. Cash flow from operations was a record $235 million.

The year finished on a high note with fourth quarter results beating our guidance. Fourth quarter revenue of $489 million increased 6% sequentially and 18% year-over-year. Semiconductor revenue was $212 million, up 8% from Q3 and ahead of our guidance as customer demand strengthened. Through solid execution, we were able to respond and capture upside.

Data center computing revenue was a record $178 million, up 4% sequentially and 101% year-over-year. Overall, demand remained very strong, and we were able to quickly adjust to meet changes in product mix within the quarter.

Industrial and Medical revenue increased 10% sequentially to $78 million and grew 2% year-over-year, the first increase in 2 years. I&M demand continues to trend positively with total backlog and distribution metrics improving over the last several quarters.

Telecom and Networking revenue was $22 million, down slightly for the quarter and the year, mainly due to program timing. Fourth quarter gross margin was 39.7%, up 60 basis points sequentially, primarily due to higher volume and favorable product mix. We continue to deliver improved gross margin despite the impact of tariffs and factory ramp costs.

Operating expenses were $107 million, up 4% from last quarter, driven by higher sales and incentive-related expenses. Operating margin for the quarter was 17.8%, up 100 basis points from last quarter and 430 basis points from last year, highlighting the leverage in our financial model.

Depreciation for the quarter was $10 million, and we achieved our second highest adjusted EBITDA of $97 million. Other income was $1.3 million, down slightly quarter-over-quarter.

Our non-GAAP tax rate for Q4 was 14.7%, below our guidance of around 17% due to favorable mix of earnings and discrete items. Fourth quarter earnings were $1.94 per share, up from $1.74 in the previous quarter and $1.30 a year ago.

Turning now to the balance sheet. Total cash increased by $33 million to $791 million with net cash of $224 million. In the fourth quarter, we delivered cash flow from continuing operations of $80 million. Inventory days came down by 3 days to 125 on higher sales and inventory turns improved to 2.9x.

Looking ahead, we expect inventory to increase to support growth in the coming quarters and for strategic supply. DSO increased to 60 days from 58 days, largely due to timing of revenue. DPO improved from 62 to 68 days. As a result, net working capital decreased sequentially from 124 to 117 days.

During the quarter, we invested $38 million in CapEx and paid $4 million in dividends. Finally, we spent $6.7 million to repurchase 33,000 shares at $205.38 per share.

Now let me review our full year 2025 results. In 2025, we delivered $1.8 billion of revenue, up 21% year-over-year. Growth was primarily driven by revenue in the data center computing market, which increased 107% year-over-year to $587 million. Semiconductor revenue increased 6% to $840 million, which was our second strongest year following the peak in 2022.

Industrial and Medical revenue decreased 11% for the full year. However, after a trough in Q1, revenue increased sequentially each quarter on improving supply-demand dynamics and lower inventories.

In 2025, we optimized our manufacturing footprint by exiting our last manufacturing facility in China, while adding new capacity in the Philippines and Mexico. In a dynamic environment, we managed the tariff impact on gross margin to less than 100 basis points.

Combined with leverage on higher revenue, gross margin improved 240 basis points to 38.7%, the highest level since 2020. Operating expenses increased 7%, well below our target of half the rate of revenue growth. Operating income increased 89% and operating margin improved 560 basis points to 15.8%, the highest level in 5 years. 2025 non-GAAP earnings increased by 73% to $6.41 per share, while adjusted EBITDA increased by 68% to $324 million.

Combined with improved days of net working capital, we achieved record operating cash flow. This cash flow funded investments in production capacity and capability to meet strong customer demand and growth ahead. As a result, 2025 CapEx was $107 million or 6% of revenue.

Turning now to our first quarter guidance. We expect Q1 revenue to be approximately $500 million, plus or minus $20 million. The sequential growth is expected to come primarily from the semiconductor market. We expect gross margin to remain around Q4 levels in the 39.5% to 40% range on similar volume.

We also expect Q1 operating expenses to be flattish quarter-over-quarter with higher investments in R&D and lower SG&A. We expect other income to be in the $1 million range and are now modeling our tax rate to be in the 16% to 17% range looking forward. As a result, we expect Q1 non-GAAP earnings to be about flat at $1.94 per share, plus or minus $0.25 on higher operating income, but a more normalized tax rate.

Due to the strong performance of our common stock and the dilutive effect of our convertible note, our non-GAAP EPS guidance is based on 39.7 million shares.

Now let me provide some concluding comments. First, we see strengthening demand across our markets in 2026. In semiconductor, we are entering the year with increased customer demand, which we expect to further strengthen in the second half.

For Data center, we expect Q1 demand to be similar to Q4 based on timing of product transitions. However, we expect revenue to strengthen through the rest of the year on higher demand and production ramp of our new programs.

Overall, we are raising our data center revenue growth outlook to more than 30%, up from 25% to 30%. In Industrial and Medical, we expect continued growth over the next several quarters on a more normalized inventories and new product adoption paced by overall economic conditions. As a result, with improved industry conditions across our markets and growth from new products, we are currently modeling high teens revenue growth for 2026.

Second, exiting 2025, we increased gross margin by 450 basis points relative to first half 2024 levels. Our initial target of 40% is within striking distance, and we expect to achieve this goal within 2026, with timing dependent on volume and product mix.

Looking forward, we believe that improved manufacturing efficiency, a growing mix of new products and higher revenue will enable us to achieve our long-term gross margin goal of 43% despite the impact of tariffs and higher data center mix.

Third, increased capital investment enabled us to double the output in the Philippines and Mexico and to complete initial fit-up of the Thailand factory. We expect 2026 CapEx will continue at or around Q4 levels, which will enable over $2.5 billion of revenue-generating capacity within our existing footprint. The complete build-out of Thailand should enable an additional $1 billion of capacity.

Longer term, we expect CapEx to revert to historical levels of around 4% of sales once we complete these investments. Lastly, our diversification strategy enables us to balance growth across our markets, generating more consistent cash flow that we can reinvest into our business. We will continue to develop power technologies that can be shared across our product portfolio and drive organic growth in each of our markets.

In addition, we will continue to look for acquisition opportunities to further expand our scope, especially in industrial and medical and leverage our scale to drive further growth in revenue and earnings. With that, operator, we'll take your questions.

[Operator Instructions] Our first question is from Brian Chin with Stifel.

Maybe firstly, I was curious, how are you thinking about your semi cap growth this year in relation to the industry WFE that is expected to grow at least in the mid-teens year-over-year kind of growth rate? And also, did I hear you correctly that you do anticipate more acceleration in the back half of the year?

Yes, Brian, this is Steve. Let me take your question. Maybe give you a more expansive answer. Maybe the first point I'd like to make is that I think we're better positioned now as a company than we have ever been in the history of Advanced Energy. And I say that largely because of the broad acceptance of eVoS, eVerest and NavX across a broad customer set. And so what that is doing is setting the company up for structural share gain over the next 5 years in three areas: in conduct, in dielectric and in deposition.

And the reason we're doing so well there is that customers are encountering problems as they go below 2-nanometer at these advanced nodes. And our technologies help solve those issues, particularly issues with throughput and yield. So if you take those design wins that we have achieved, couple that with increased etch and depth intensity at the leading edge, I think it sets up very nicely for share gains for Advanced Energy in the semiconductor area.

But we have some other factors that are in our favor as well. One is a larger installed base of AE boxes that are installed in fabs. And what we're seeing right now is a surge in demand for advanced logic capacity as well as DRAM capacity. And since we're strong in conductor etch, we expect to see a fair amount of business as those capacity build-outs take place in the coming 12 to 24 months. So we think we're in good shape there. Our service business continues to grow. We have a large number of boxes installed throughout the world there that require service and other added value functions.

A new area for us is system power. We had pretty much ignored this in the past. But over the past few years, we have focused on system power solutions for semi equipment and semi tester companies. And we've achieved some success there. Those programs will ramp in '26. So there's a lot of growth factors in play for us. We don't know exactly how much we're going to grow in '26, but I think we'll be happy with the growth once the dust settles.

Okay. Appreciate that. It sounds like some of those design wins maybe start to give you some visibility there in the second half.

On the data center, maybe to switch over there. I guess, can you discuss what you're embedding in your greater than 30% revised growth outlook in terms of new customers? And then second part of that, for existing customers, are you pretty optimistic that volumes and activity should further strengthen as we go through the year here, given that the sheer magnitude of CapEx increase that big hyperscalers are guiding to, which I think is sort of like a 70% increase or so on average?

Yes. So let me answer the question about the forecast. Our forecast of over 30% growth this year only comprehends our existing customer base. It does not comprehend any pull-ins of demand from second wave customers.

But what we're seeing right now in the market is pretty bullish. I think you've seen the announcements from the hyperscalers about their capital spending plans, which are up to the right. We've seen very bullish forecast as well. So we're preparing for a strong year in '26 to meet our existing customer demands.

To that end, we have spent a fair amount of money, as Paul described, on expanding capacity in the Philippines and in Mexico to support this growth in demand from the hyperscalers.

In addition, we have brought Thailand to a place where we can basically start that factory up this year, if need be, and absorb any demand that we can absorb in Mexico or the Philippines. So I think we're in very good shape from a capacity standpoint. I think the other positive is the activity on the development side. We're fully engaged with our customers, including a number of 800-volt projects. And for us, rapid change is a good thing because we are technology leaders, and it makes it harder for the other guys to catch up essentially. So we're pretty excited about data center in '26.

Our next question is from Krish Sankar with TD Cowen.

Steve, just to follow up on the previous question on data center. I'm curious, what is your visibility into these projects? Because clearly, at over 30% growth, it seems like that basically implies mid-single-digit growth sequentially from Q2 onwards. But you've kind of healthily overgrown that number over the last several quarters. So I'm just kind of curious that greater than 30%, is that conservatism baked in? Or is it more lack of visibility into the projects? And then I had a follow-up.

I would say this. I think there's be upside to our number. I think one of the issues we face, Krish, is on the supply side. And we've seen this play out in '25, and we expect to see more of it in '26. And I think what we've seen in '25 are various constraints on processors, whether it's GPUs or ASICs or some other type of processor product. And that has dictated in some cases, the number of boxes that we need to deliver to our customers.

I think as we move into '26, there's additional constraints, namely in memory, right? I think most of the memory makers have announced that they're sold out. And so you've got allocation situations in both processors and memory, which will limit some of the growth, we believe, in 2026.

So that's why we're a little conservative. That said, supply chain issues have not limited our ability to build products for the data center and certainly not in the first quarter of '26. We do anticipate that we'll see some supply chain issues moving forward. So we're putting our thumb on the scale when it comes to building inventory. We're trying to put strategic inventory in place where we see weaknesses in the supply chain.

Got it. Got it. Very helpful, Steve. And then just a quick follow-up on the semi side. You mentioned second half better. Is that just more a revenue commentary for you for your semiconductor sales? Or is that more an inflection commentary? I'm just wondering because if you look at [ ICO ] last night, they're beginning to see an inflection in the semiconductor sales to the semi-cap OEMs already happening. So I'm just wondering that since you're a supplier of components, you should start seeing it either in the March or the June quarter. So I'm just wondering, is the second half commentary for semis more just half over half revenue? Or is it more of an inflection commentary?

Yes. I think we saw a material change in customer outlooks in Q4. And that's what drove our Q4 outperformance we saw that demand go up. Originally, we were concerned it might be a pull-in from Q1, but then we've discovered now the Q1 demand also went up as did Q2. Our customers have also told us to expect further increases for the second half. And so we're getting a lot of positive comments from our customers and improved forecast, which lead us to believe that the second half will be stronger than first half, and that will lead into a pretty healthy '27 as well.

Our next question is from Mehdi Hosseini with SFG.

I have two. The first one for Steve. As you ramp the Thailand facility and get it to revenues of $3.5 billion, how should we think about that revenue mix between semi, data center and the rest?

Yes. So just to review, what we said today was that we would have more than $2.5 billion in revenue-generating capacity in our existing factory network. And then we brought on Thailand that would add another $1 billion or more so that by the end of next year, I think you're looking at $3.5 billion, assuming we build out Thailand. So it's more than enough capacity to achieve our goals over the next couple of years. So that's the important thing.

The exact mix, we haven't really gone into that. I think we're seeing growth in all of our markets. I think initially in Thailand, we're looking at data center because it's a high-volume, low mix type of product. It's probably the best way to start up a factory. But I think we're also going to see our plasma power products being built in Thailand in the near future as well. And I think the third category of products that will go into Thailand will be the industrial medical products. But that -- this is our biggest factory. It's 0.5 million square feet, and it was built to accommodate all of our products. So it's the first time we've had such a factory in our network.

So should I assume that if on the semi side, the market were to exceed well over $150 billion, you have no constrained capacity. You can reallocate internal capacity to meet your OEMs demand above and beyond the $150 billion WFE?

Yes, definitely. In fact, that was the original impetus for us to build Thailand was as a business continuity factory for semiconductor. And so our customers are fully bought into Thailand as a second factory to back up our operation in Malaysia.

Great. And my second question has to do with the data center and migration to 800 volt. I'm under assumption that there would need to be redesigned. And what I wanted to better understand from you, would there be ASP uplift for premium associated with the redesign of these power sources and you're also increasing your content. So I know some of my peers are fixated with data center CapEx, but I want to better understand the details. So the question is, would there be an ASP uplift as you start supporting an 800-volt AI data center rack?

Yes, interesting question, and we've spent a fair amount of time looking into that. And we're engaged with multiple I would call them marquee customers on 800 volts because we have some pretty interesting technology that makes 800-volt possible in a relatively small space.

So I would say that based on our analysis of the market, our total dollar opportunity goes up with 800-volt solutions relative to what we're generating today. So in addition to having the right technology, I think it's also good for our business to mix some of these newer technologies into the mix. It's very good for Advanced Energy.

Our next question is from Steve Barger with KeyBanc Capital Markets.

Steve, for data center, your comments on processor and memory constraints make sense. But if you grow the 30% for existing customers and the second tier does come in, do you have capacity now to support upwards of 50% growth? Or what can your factory support if everyone else in the supply chain delivers?

Well, that would be a great situation, Steve. I hope you're right. But that's the reason we built Thailand, right? So we will likely qualify our first products on Thailand this year. And so -- and the purpose of that would be to be able to ramp those products as soon as Q4 of this year.

And that would likely be data center, whether it's with our existing hyperscale customers or some new second wave customers, that remains to be seen. But I don't see factory floor space or equipment being constraints for Advanced Energy. I think our attention is more focused on bill of materials and parts and ICs and discretes and those types of items. And so that's why you're seeing our inventory move up a little bit because we're trying to take some insurance so that we don't get caught short on the parts.

Yes, understandable. In semi equipment and your comments on really strong customer forecast and a strong back half, what are they saying about their desire to hold buffer inventory? Is there some restocking built into your expectation? Or are you just currently selling through to current demand?

It's a difficult question to answer precisely, Steve. But I could tell you, I think for much of the past 2 years, the demand has been pretty close to equilibrium. So they're ordering what they need, keeping reasonable inventories. It's been pretty stable, I would say.

I think there's certainly a change as we move into '26 and '27, where our customers see more upside from their customers. And so yes, I think that we'll see them holding more safety stock because we all have memories of what happened back in '21 with some of the supply chain shortages, and we don't want that to happen again.

Our next question is from Joe Quatrochi with Wells Fargo.

Maybe just kind of in that line of questioning, I guess, then as I think about -- when we think about just your semiconductor equipment growth kind of opportunity for '26 and then you layer on top of that, the new products, I mean, why are we not to consider that, that business could grow upwards of 20% this year?

Yes, Joe, I think it could. I think there's an upside to what our forecast is in right now. But we're not ready to forecast that. We just want to see how the market develops. There's a lot of timing issues with wafer fabs and exactly when the clean room space is available to accommodate new equipment. So I think that will all become clear in the coming months.

Okay. And then as a follow-up, I'm wondering if you could just kind of give us maybe the puts and takes of the gross margin guide. I guess I would have thought it maybe be a little bit better just given it seems like mix be a little bit stronger in semi might have been a benefit for this quarter?

Yes. So we did see gross margins increase this quarter from last quarter by 60 basis points or so. The one thing I would note that if you recall, we had a tariff headwind. We had pretty favorable tariffs in Q3. So that was pretty significant that we overcame that.

But you're right, we had a little bit better mix certainly and a little bit better -- higher volume, which we're able to absorb. So we were pleased with where we ended up. I would say if you excluded the impact of tariffs, we are clearly well over 40% gross margin, which was certainly a goal exiting this year.

I think if you look forward, we see gross margins flattish in Q1, largely on a pretty similar mix. Revenue is up a little bit. But I think the opportunities for us in gross margin come to continuing to improve manufacturing efficiency. We are still carrying a fair amount of ramp costs as we ramped up data center. And the mix in that business continues to be pretty dynamic, including in the fourth quarter, and we expect some of that in Q1 as we go through some product transitions there.

So we certainly see opportunities to improve gross margins. We're very comfortable or confident that we'll see over 40% in 2026. And when we look forward and we have more tailwind from mix within the new products that we're bringing out. Certainly, semi mix will help us. even in Industrial and Medical, there's opportunities to see improvements as our new products play a bigger role to see gross margins continue to improve. And we said in our prepared remarks that we believe we still have line of sight to 43%, which is our long-term goal as volumes grow, as our mix improves and as we continue to improve our manufacturing efficiency.

Our next question is from Jim Ricchiuti with Needham & Company.

And Paul, just a follow-up on the gross margin question. As you -- with the continued rapid growth in the data center business, I'm just wondering how we should be thinking about gross margins in that area of the business. Can we see further improvement from here? Or to what extent does mix play a factor in this?

Well, mix will play a factor, but it plays a less -- a smaller factor than certainly it has historically. And I think if you look at 2025, in particular, we grew gross margins pretty significantly despite our data center mix growing from sort of the low 20s to close to 40% for -- by the time we exited the year.

As data center continues to grow, there could be some headwind there, but the margins there continue to approach corporate average. And certainly, as we bring out new products in data center and we become more efficient in manufacturing, we think we can largely offset the impact of that mix. So we're not backing off our goals overall based on what we see today. And we believe we can largely accommodate that within the sort of plus or minus 50 basis points kind of net mix impact that we've discussed.

How could you say what the new products of eVoS and contributed in 2025? And I'm just wondering, Steve, when you talk about potential upside in the semi business in '26, that do you see the new products being a big driver to that or just strength in the existing portfolio?

Yes, Jim, I'll take a cut at that, and then maybe I'll let Steve give some color. But we did meet the goals that we laid out for those new products, seeing double-digit millions of revenue for those new products. Remember, these are largely still in the qualification and what I'll say, early pilot or early production stage. And the ramp of these is largely tied to sub-2-nanometer process ramps.

So we're excited because we're seeing a lot of qualifications. We're seeing pull for the products to get in those qualifications. And we would expect to see revenues in '26 higher than '25 certainly. And as we see production ramp of these next-generation nodes and those get on more and more process steps is when we'll see really the pull-through of these new products.

Yes. Jim, let me just add a few comments. I think one of the interesting things that we're seeing is as customers realize the benefits of these new eVerest and eVoS and NavX products at the leading edge. They're starting to think about incorporating those products at non-leading-edge processes.

So once we get into a customer, show what we're capable of with this technology, we're seeing a lot more opportunities beyond the initial opportunity. So I think what we're going to see in the coming years, maybe not in '26, but in '27, '28, '29, we're just going to see the usage of these products multiply, and it's going to be basically a share gain driver for the company.

Got it. And one final question. I didn't hear a whole lot about the M&A pipeline. I'm just wondering, you guys got a lot of things going on. How active is the pipeline right now?

Yes. We're still on the hunt, and I would characterize the pipeline as active. We did do an acquisition of Airity not so long ago that the technology from that company has played a critical role in some of our new products. So I'd characterize that as a success. But we also see opportunities on the industrial medical front. And we think with the market normalizing, it will be easier to reach agreement with potential targets on the valuation of their assets. So we're optimistic.

Our next question is from Rob Mason with Baird.

I was curious, Steve, around the second wave of data center customers as those customers begin to ramp, how would -- what would you view the gating factors around those ramps and maybe even the potential pull-in if that was to happen around that set of customers, what would be some of the major influences -- would some of the supply constraints you mentioned have an outsized impact on those customers' ability to ramp?

Yes. So one of the things we like about the second wave customers is they don't require a lot of engineering work on our part. So with the hyperscalers, it requires a lot of work. So that's where we spend most of our time on the engineering front.

I think you're correct. I think as the contention for processors and memories heats up, I think the hyperscalers have an advantage there. And so it may impact some of the ramps of the second wave customers. That said, I don't know that for a fact, but I would think it's going to be somewhat challenging this year to get as much memory and processors as you really want.

Understood. And just as a follow-up, maybe a point of clarification. Paul, you talked about industrial and medical continuing to grow for the next couple of quarters. I just want to confirm that that's off the fourth quarter as kind of a jumping off point. You were inferring growing sequentially?

Yes. I think if you look at our guidance on balance, we said the Q1 growth would be driven primarily by semiconductor. So that infers kind of a flattish quarter generally for Industrial and Medical and data center, but for different reasons.

In Industrial and Medical, we're seeing all the vectors point the right way. So we're encouraged about that. We also acknowledge there's typically a little bit of Q1 seasonality. So we think sort of flattish would actually be sort of improvement, if you will, all things else being held equal. And then certainly, we would expect to see growth over the course of the year as those vectors continue to improve.

The one thing that's the wildcard there is just the broader macro economy. There are certainly some sectors in industrial and medical that are doing better than others. I think if the economy stays steady and maybe some of the uncertainty comes out of it, we could see that growth accelerate.

In data center, I think we talked about flattish in Q1, mainly on product transitions as new products come in, get qualified, displace the older products, we expect to see a flattish quarter with then solid demand and ramp after that in that market.

Our next question is from Scott Graham with Seaport Research Partners.

Congratulations on your print and your high teens sales thinking for '26. That's pretty neat. I wanted to understand a little bit more about a comment you made, Steve, about market growth in Industrial and Medical. You said you're positioned to outperform the market. What is the market growth in your definition in those businesses?

Yes. So the reason I said that Advanced Energy is positioned to grow and to outgrow the market is the investments we've made over the past 4 years. And so despite the fact the market was down, we continue to pour money into new product development, into digital marketing, into channel development in all the areas that you need to invest in to grow your business. And so what that created was a very healthy design win pipeline. And it takes a few years on average for these design wins to go to revenue.

And so I think based on the design win pipeline and what we're seeing right now that we can grow faster than the I&M market in '26 and '27 and '28, too. But we're going to keep growing market share in that market because we have invested heavily and we've been successful in winning some pretty high-volume designs.

Understood. This follow-up question is with Thailand coming up the capacity curve here and equipment being installed, I'm curious that your gross margin aspirations don't appear to have changed. I'm wondering though, if operating expenses, which -- where your goal is to increase those by half of sales or less. Is that still doable in 2026 with Thailand coming online?

Yes. I think they're kind of separate. The vast majority of the costs for Thailand are going to be in cost of sales. Certainly, there's going to be administration. We're going to have accounting people there and other things that would fall into OpEx. That is contemplated in our projected growth.

And maybe just to clarify a couple of things. So first of all, on the margins, the Thailand factory has always been contemplated in our 43% goal. So that's not new. Basically, the question was when do we have the volume and the demand to support it. And with the growth in data center and now semiconductor really heating up, that growth -- that time frame is pulling in. So we're really glad we made those investments, and we're positioned to be able to fold that into the portfolio.

On the OpEx side, as we mentioned, we performed really well in 2025. I think we grew OpEx only 7% and revenue over 20%. So that's well ahead of our model.

If we look at 2026, we'd expect to continue to make investments. spending is about flat in Q1. But after that, we'll have salary increases, targeted investments that we're making, including in places like Thailand. There'll be variable costs that go with revenue growth. So we do project that we're probably going to see OpEx grow through the year probably to an exit rate of around $120 million by the end of the -- by the fourth quarter, which probably puts OpEx in the $450 million to $460 million range. And we do think we can accommodate the outfit of Thailand in the operating expense envelope for what would be in there.

[Operator Instructions] our next question is from David Duley with Steelhead Securities.

I was wondering, typically, at the beginning of a semi equipment ramp, you would grow a little bit faster than your big customers just because they're going to replenish inventory. Do you think that's the case this time?

Well, I think if we look at '25, we grew semiconductor 6%. We had our second highest revenue year in our history. So I think our customers had a decent amount of inventory in place going into '26. But I think what we've seen is an acceleration in demand. We saw that in Q4. We'll see -- we're seeing it in Q1, and we'll see it throughout 2026.

I think at the end of the year, we'll take a look and see if we've grown faster than the market, but it's pretty difficult to tell right now exactly what's going to happen from a demand standpoint.

I'll just add to that. I mean it's just -- it's short memories. But if you recall, we expected Q4 to be a down quarter based on everything our customers were telling us. At the end of Q3, we expect to be down 2% or 3%. We actually ended up growing 8%. So that's a 10-point swing.

So I think we're already seeing a little bit of that, David, and we'll see how that progresses through the year. As Steve said earlier, our confidence around strong -- really strong second half continues to grow based on the signals that we've seen from our customers, and we'll see how the timing plays out.

Yes. I think in your previous slide decks, you've talked about being able to outgrow the WFE market. Whatever the market growth is, you can grow, I think, by 30% more. And I guess whatever the WFE market growth rate or size tends to -- is going to be in '26, do you think you can outgrow that market growth?

Yes. I think there are a lot of issues that determine your growth from quarter-to-quarter, from year-to-year. So what we do is we take a look at a little bit longer time frame. So in our case, we look at 3-year and 5-year CAGR essentially. And when we look at that versus our peers, yes, we're definitely growing significantly faster than WFE. But there are definitely variations from year-to-year where we're going to be either much better or a little bit worse than others, but that's due to tactical factors.

Okay. And just a clarification from me. Paul, I think you mentioned in your prepared remarks that you were going to strategically increase inventory levels just to make sure that you can meet future demand from your customers. What sort of increase in inventory should we expect over whatever time frame that you're referring to?

Yes, it depends on where revenues come out, David, but we exited the year around 2.9 turns. I think we'll probably see a little bit of a haircut on that at the beginning of the year, maybe down 1/10 of a turn or 2, but then recapturing that as time goes on and revenues grow. I think the key thing is we want to make sure we've got the strategic parts. So that's not buying everything, but it's making sure we're looking at items that have been problematic in the past to make sure we've got good strategic levels of inventory. And we're positioning ourselves for, as Steve commented on the ramp that our customers have talked about is coming.

Okay. Yes. So it sounds like -- and I think you already said this, but I just want to double check, you're not going to be the gating factor in your customers' ramps?

That's right.

No, we don't expect to be the gating factor.

Thank you. There are no further questions at this time. This does conclude today's conference call. You may disconnect your lines at this time. We thank you again for your participation.

I guess we can get started. Hi. I'm Shane Brett, U.S. semiconductor equipment analyst. Joining me today from Advanced Energy are Steve Kelley, President and CEO; and Paul Oldham, EVP and CFO. Before I start my questions, I'm going to hand it over to Paul for a quick Reg FD disclosure. And for Steve, if you want to give kind of an elevator pitch of what Advanced Energy does after that.

Yes. Thanks a lot, Shane. Thanks for joining us today. Just a reminder that any comments we make today may be subject to certain risk factors. You can find a better discussion of those in our SEC filings. Also a reminder that we are not providing any update to guidance today. Our earnings release was on November 4. So please refer to that as our latest guidance.

Okay. Maybe a quick overview of Advanced Energy. We're headquartered in Denver, Colorado. We're about a 45-year-old company. We got our start providing power solutions for etch chamber makers, so for Applied Materials, Lam and others in that business. So about half of our business today is in semiconductor equipment. We provide these critical subsystems.

The other half is what we call system power. So most of that business is for AI data center, but we're also a significant player in industrial and medical products. And in all the markets we play in the high end, over 70% of our revenue comes from sole-source products. So we're on a journey to get that above 80%. And our goal over the next years, as articulated in our Investor Day last November, is to double the size of the company to $3 billion in revenue and about $15 in earnings per share.

Great. As a semiconductor equipment analyst, I've always thought of you guys as a semiconductor equipment company, but I think you're probably a data center company now or -- I just want to kind of talk about the data center business as you're on track to double it this year. Just what's driven this growth? Can you give us kind of a bit of color on what's the drivers and what's kind of the visibility into '26 on data center?

Yes. So when we bought Artisan back in 2019, they were in the data center business, but they were supplying commodity solutions. So we faced the decision roughly 3 to 4 years ago about whether to exit the business because of the low margins or to reorient the business. So we chose to stay in the business and focus our engineering team on higher-end solutions.

So trying to tackle the most difficult problems. And so that strategy has worked out well for us. Today, the margins we generate in the data center business are just below corporate average, which is a huge increase over what we used to be able to generate. We're engaged with a select group of hyperscale customers. We're going to grow over 100% this year relative to revenue in 2024. And we forecasted to grow 25% to 30% in '26. And this is solely with our first wave customers, the select group of hyperscalers that we've engaged with.

We're also starting discussions with second wave customers, which don't require the same level of engineering intensity. We could modify basically standard solutions to meet their needs. So there's some upside looking forward. We think most of that upside occurs in '27.

I'd say 100% growth is probably a little bit better than going well, but -- yes, it is going well.

We're pleased.

Yes. So I guess there's just so many data center projects going on right now. And I think just within Morgan Stanley researchers seem to consistently be raising numbers. But where are you guys in terms of meeting this data center demand? And how big of a challenge has it been to sort of expand capacity to meet this demand?

Yes. So there's basically 2 areas where we need to invest. One is in development and the second is in production capacity. So development is largely a function of our engineering team. It's also a function of our investment because these higher power products require more expensive labs and development equipment. So we've invested a lot.

We've invested much more than we normally do in a year. Actually, the last couple of years, we've overinvested in CapEx, but -- and most of that's gone for data center capacity and development products. So we have kept up with demand. We have expanded capacity in the Philippines and Mexico. And we have a third factory will likely ramp next year for data center in Thailand.

But in terms of this new capacity there, like the payback period isn't a matter of years. It's almost a month.

Yes. So typically, our payback on this new investment is 9 months or shorter. So it's a quick payback. And the equipment is fungible, which means we could use it for other products such as semiconductor, industrial and medical.

Got it. I just want to kind of follow up on that 25% to 30% growth you kind of talked about for 2026. Granted you are a sole-source business, I think the hyperscalers will probably give you a certain amount of visibility. How much does that -- how much does your visibility from these customers extend? And maybe how early are you kind of brought into new projects with these hyperscalers?

Yes. So our visibility, I would say, is roughly 9 months. We get a forecast and sometimes a PO for 9 months. So the visibility is good. We need to be involved with our customers on these designs very early because what we're seeing now is a yearly cadence of new products and a lot of it is tied into the introduction of new GPUs. And so for every new GPU, you typically need a new power solution because each GPU typically uses more power.

And so today, we have secured the design wins necessary to support our forecast for '26. So what we're actually working on with customers today are designs for '27 and '28. And so that's really the cadence is we're looking at not M+1, but M+ 2 and beyond with our customers.

Got it. Got it. I guess a little bit more of a longer-term question, which may be interpreted as a short-term question. But how would you think about the longer-term growth prospects of this business? And is there sort of any way you've thought about your attach rate sort of revenue opportunity per gigawatt or like dollar of hyperscaler CapEx that you've sort of thought about?

Yes, it's difficult to identify an attach rate or some growth with the industry because if you think about it, our whole strategy is to be a specialty player. It's to focus where there's unique opportunities for customization where we can bring great technology to market.

I think what we could say, though, is that as this data center market evolves and continues to require more power and more capability that our content at each of those levels should continue to increase.

Got it. And would there be a bit of a content story with these kind of second wave of customers that you've identified?

Yes, absolutely. If you take a look at what we've done is we've developed the core technology building blocks, if you will, to serve our primary hyperscale customers. Now we're seeing the second wave of customers come in, the so-called Neo cloud, CSPs or even enterprise customers who are looking now for more rack level solutions. We can repurpose those technology blocks to their needs with just a little bit of customization.

And so we get good leverage, we think, on the engineering investment that we've made. We don't have to create all new solutions. And we can bring very good products to market that are proven, reliable and leading edge to this next wave of customers.

Got it. Got it. Now I'm going to move over to semi, which I'm probably a little bit more comfortable with. So you've had a nice year in semi. Q4 guidance implies about low teens growth for the full year. Can you talk about what kind of -- what's driven that? And your customers have talked about a really big -- well, not really big, but a good second half 2026 ramp. Just what are you doing in terms of preparing for what the customers have kind of talked about publicly?

Yes. I think the real story in semiconductor is our success with the new products because that's laying the groundwork for sustained share gain moving forward. So in '23, we introduced 2 new platforms called eVerest and eVoS together with an accompanying product called NavX. And we've had great success on the conductor etch side of the business. So you're going to start seeing that in the form of significant revenue next year.

We're also having success in the dielectric etch part of the business, and that's a new area for us actually. And so we anticipate significant revenue starting in 2027 in that area. So we're expanding our TAM, and we're going to grow our market share over the next 5 years based on the success of our new products. I think short term, all of our customers are saying similar things about second half strengthening next year and 2027 being a really good year. And I think that's based on a number of factors, including when new fabs are coming online.

Got it. Yes. So as we kind of think about we're headed hopefully towards a pretty good '26 and '27 for WFE, you guys have laid out a goal where I think it was 20% or 1.2x or 1.3x above WFE. How much of that can we kind of attribute to these new products? And how much of that do you think we should think about in terms of just etch dep intensity increasing within WFE?

Yes, it's a little bit of both. I think our customers have been vocal about the fact that etch and depth intensity is going up. And that's the nature of these new processes, just a lot more steps, a lot more etch and deposition going on.

So that's good. It's good for our customers. It's good for us. In addition, we think we get a bump from our success with the new products. So you're going to see eVerest and eVoS in many of these new processes that are ramping over the next 2 or 3 years. And that's both in logic and memory applications.

Got it. And I guess between logic and memory, how would you sort of characterize your exposure? Would you characterize -- how would you characterize sort of your excitement levels for the 2 end markets into '26?

I think at the leading edge, it's pretty exciting across the board. I think certainly for leading-edge logic, the AI demand is driving these new process developments. It's driving volume. And obviously, the leader there is TSMC, and they're doing quite well. On the memory side, very exciting as well with high-bandwidth memory, DRAM and NAND.

Also, I think all 3 of those markets will be strong next year, probably into '27. So I think we've turned the corner on leading edge in semiconductor. We're pretty excited about our ability to grow share in all these segments.

And I want to go back to eVerest and eVoS. But as sort of your main customers come over to you, what are they -- how much -- how demanding are these customers in terms of asking you for R&D, getting you to sort of kind of enable the next wave of leading-edge semiconductors?

Yes. I think semiconductor customers are among the most demanding, right? They're very focused on developing solutions that meet the needs of their customers, the fab operators. And so it's an intense process. I said we launched the platforms back in '23.

It typically takes 2, maybe 2.5 years to customize the product, and it's a 3-party development. It's us, it's our customers and it's their customers. And so we basically tune our subsystems to their chambers and then work with them to tune it further to the end customer processes.

Got it. And just last, a little bit more shorter-term question. How would you characterize sort of the inventory situation at your customers? Or just given the customer concentration for semis, is that not something we probably need to think about too much?

Yes. I think the inventory situation is healthy. I wouldn't say there's too little or too much. It's about where it should be. So we deal with our major semiconductor customers on a JIT basis. And so we basically fill the bins to wherever they need them fill to.

And that's going to vary depending on their view of the market, but we're keeping up with demand. And we have upside capacity at our factory in Malaysia. And ultimately, we're going to qualify the semi products in both Malaysia and Thailand. We have 2 factories.

And you're speaking with these customers every week, so you're on top of it.

Yes.

Got it. So I want to move over to industrial and medical. And I guess, to be a little bit critical, data center has absolutely exploded. Industrial, medical has been a little bit lackluster. But could you talk about what you see in that what you're seeing in that end market, what you're seeing in that business segment and how you're kind of thinking about the 2026 setup for here?

Yes. I think Industrial Medical has been in a correction mode for almost 2 years. And so I think the industrial medical customers, many of them suffered with the COVID supply chain shortages, and it's taken them a while to work through their inventories. I think today, most customers have worked through the inventory.

The rest of them should be finished in the next 3 to 6 months. So we're seeing incrementally improving business in '25. We hit bottom in Q1 of '25. And each quarter since we've been a little bit higher in that business, a little higher revenue.

We see some good trends in distribution, where the inventories have declined 7 quarters in a row, and we're at equilibrium now in distribution. So we think in Industrial Medical, '26 will be a growth year. It's not going to be a V-shaped recovery. We think it's going to be gradual. But we have a very solid design win pipeline, and we think we could start showing outsized revenue growth in '26 and '27.

I guess to ask a deeper question. So I feel like in semiconductors, when we're in an up cycle, we call it secular. When we're in a down cycle, we call semiconductor cyclical. But just how does this industrial and medical portfolio play a part within the Advanced Energy portfolio, just given how cyclical in hindsight, semiconductors and data center may be?

Yes. I think industrial and medical plays an important part. So the way we think about the business, we have 3 pillars. We have semiconductor, we have data center, we have industrial medical. And at any one point in time, at least one of those markets will be up. And so we can sustain our R&D spending, our capacity investments because under any condition, we're generating good cash flow as a company.

So we think it strengthens us as a company, enables us to create some space between us and our competitors. That's the overall view. Industrial Medical is the weakest of 3 pillars. So part of that's market-based. Part of it is our starting position.

So moving forward, our primary focus for M&A will be adding more to our I&M portfolio. In addition, we've been investing for the last 3 years in strengthening our channel and strengthening our direct sales force calling on industrial medical accounts. So we think the organic efforts, together with some inorganic efforts will boost us to a #1 position in industrial medical.

Got it. I want to move over to some financial questions, and then I'll open it up to the audience. But -- so you're on track to about, I think, low to mid- $6 of EPS this year, in line with your sort of $1.75 billion revenue model. Assuming we get a good year in semi and data center, it doesn't feel like we're too far away from you guys' $2.5 billion target you laid out for 2030, but it feels like it's not too far away. Just not asking you to update your guidance, but just relative to when you kind of put out this 2030 model, how different is the operating environment been for you guys?

Yes. I think the biggest change that we've talked about it is really the explosion of the data center market. And that's really come on seeing, as you've seen, all of AI explode, combined with our change of strategy Steve mentioned earlier about targeting specific areas which we've been able to win. So that market is running well ahead of what we thought.

And in fact, we'll probably achieve our 2030 target that we set a year ago in data center either this year or early next year. So we're well on track there, a little bit ahead. And as we said, as we look forward, we think that's sustainable at these levels or higher, at least over the next year or so with reason to believe with the second wave of customers that we can sustain that and even grow that beyond that point. Semiconductor is -- this 2025 is a good year for us.

We'll grow, and it will be our second best semiconductor year ever. Having said that, if you go back a year ago, the WFE projections for growth sound a lot like they sound now. I think the difference is now I think there's a lot more end market data points to corroborate the 2026 second half, we'll see growth. So I think we're generally on track there if you looked at a post '26, early '27 level.

Industrial and Medical, as you pointed out, frankly, was a little softer. Our trough level in Q1 of this year was lower than we expected, but we're slowly building out of it. So we feel confident in the model that we are on track from a revenue perspective to achieve $3 billion, and we could achieve it early. I think it's very possible depending on how the markets go. I'll also just add that we feel like we're well on track from a gross margin perspective.

I guess I want to touch on that margin portion because I guess the next step for you guys will be the $2.25 billion level, which is 30% growth from here. But when I think about that, your growth -- gross profit margin should be 42%, operating profit margin, 20%.

I feel like you've also been spending a lot more on CapEx just given where end demand is. Can you kind of talk about how that sort of increased CapEx or sort of a changing end market mix per se may sort of impact that financial model that you are envisioning?

Yes, it's a really good question. When we look at gross margins, we have improved them almost 400 basis points over the last 5 quarters and 280 basis points just last quarter alone year-over-year. So we think we're on track. We delivered about half the gross margin improvement to get to the 43% that we talked about earlier.

From a CapEx perspective and the additional headwind that goes with that, I'm not too worried about that because that CapEx was contemplated in our model. Bringing on Thailand was always contemplated. It was contemplated at certain revenue levels. We're achieving those revenue levels earlier. So I think that works well within the model.

Frankly, what wasn't contemplated was the tariff environment, which has been about 100 basis points of headwind for us. And what wasn't contemplated was the high level of mix of data center at this point. Now semi will recover, and I think that mix will normalize a little bit over the next couple of years. Having said that, we believe we can still achieve the 43%. There's other levers we can pull operationally to help offset the tariff impact if those don't get mitigated hopefully on their own.

And we see opportunities, frankly, to continue to improve overall margins and absorb the data center mix. Frankly, we think it's pretty positive that we've absorbed a roughly doubling of our data center business, and we're still on track to get close to 40% exiting this year or early next year.

Got it. So can I -- so when I think about the 3 drivers of gross margin improvement, which is manufacturing and cost improvements, just product mix and then volume leverage, would you say that kind of the contribution from them has been a little bit more skewed towards the volume leverage over the last year or so?

Well, I think the first thing is we've executed on our factory consolidation plan. So we talked about 200 to 250 basis points of improvement on that. We've largely achieved that, not completely, but largely. And like I said, I think we can actually do more there. So over time, we'll get a little bit more. We've been right on track with volume. So certainly, volumes up. That's fit right with our model of how we would expect that to occur.

The third leg of margin improvement that we haven't realized as much yet is the portfolio mix. And portfolio mix has mostly to do with as our new products take hold in the market and become a bigger portion of our revenue, those come with higher margins. And we think there's 200 to 300 basis points of improvement that relate to that. Now that won't happen immediately because that products have to get in the market, they have to ramp to scale.

But we feel very good about that aspect that, that will occur over time. So on balance, we feel good about the strategy and the execution that we've had and think we'll still be able to get to the 43% ultimately and overcome some of the headwinds we have.

Got it. And just one more question for me, and then I'll pass it over to the audience. But just I&M and sort of how M&A sort of kind of plays into the growth strategy there?

Yes. So if you look at I&M, Industrial Medical, it's a highly fragmented market. So we think there's an opportunity to do a partial roll-up. We play in the high end of the I&M market. Let's say, the top 1/3 of the market is where we play. So we're looking for similar companies that are out there who provide these industrial medical products, which are sticky, right?

Once they're designed in, they stay designed in for a very long time. They tend to be long life cycle, 10, 15, 20 years, and they tend to be very profitable. So that's our kind of business. It's hard to displace existing competitors. So you have to go out and do some M&A to really increase market share over the short to medium term.

With that, I'd like to open up to the audience if there's any questions. If not, I'll have plenty of questions about semiconductors, but -- Damn. Actually, I want to ask about data center there. So this 25% to 30% growth for next year, how would you sort of characterize the risk to the upside, just given 100% was probably maybe a little bit higher than what you guys are envisioning this year? Just how would you characterize that risk to the upside?

So risk to the upside, does that mean is there an upside?

Yes.

I think there are opportunities. We charted out '26 based on the design wins we have with our Tier 1 customers. We have not included Tier 2 customer opportunities in '26. We have those penciled in for '27 revenue. But is there a potential to pull it into '26? Yes. but we're not at that point where we could confirm it's going to happen. So I think the bias is probably towards the upside in '26 for data center.

But yes, would that be an Advanced Energy capacity issue or Advanced Energy is being a little bit more selective on, okay, these are the projects that we'd want to go after and kind of prioritize?

Yes, definitely not a capacity issue. We've invested a lot. So we have sufficient capacity in the Philippines and in Mexicali today. We can also ramp up Thailand next year. And so that factory is completed. And for about 6 months from a go signal, we could start to produce product in that factory. So we've got substantial upside, I think, in the second half of next year if we need it.

Got it. And...

What's the risk of double ordering.

So just to repeat, the risk of double ordering in data center and data center?

Yes.

Yes. The risk of double ordering seems to be minimal. In fact, as a company, we really haven't experienced double ordering, I think, over the last 5 years. Most of our products are sole source. And so customers order them when they need the products. And so whether it's data center, semiconductor, industrial, medical, we haven't seen double ordering. So it's not a risk for us that I'm aware of.

What's the lifetime of your products, especially in data center?

Yes. So the question is what's the lifetime of our products, particularly in data center. The lifetime is quite long. We build highly reliable products. So I would say, typically, they're going to last for 7 to 10 years. So it depends on the price of the product.

Ultimately, the customer can choose to replace our product with a new box. But in the semiconductor market, the pricing is so high that we have a pretty thriving service business where periodically, we bring the boxes back into our shop and calibrate them, repair them, upgrade them. And it's a very good business for the company. But that's primarily semiconductor.

How much part of semi business is the service part?

Service overall is about 10% of total revenues. And so within semiconductor, you could think of it in the mid- to high teens as a percentage of revenue. So it's healthy, and it's been growing and a lot less volatile. Much more stable, yes, absolutely.

I guess going back to that sort of data center product lifetime question. But I would say your kind of customers are constantly asking for new products and innovation there.

Yes. So basically, it's a continuous process. We're innovating every year. So the customers are trying to keep pace with the introduction of new GPUs and equivalents. And so they bring us in early in the design process so that our power solution is ready and it's an integral part of their rack design.

So the hyperscalers are demanding customers?

Yes. So we have a lot of demanding customers, semiconductor and hyperscalers top the list.

Got it. Still have 30 seconds left, but there are no questions, I think we can wrap that up for today -- Oh, sorry.

I mean I can say that the next 2 years, data center could be strong, but I mean, no one expects the amount of CapEx to be the same in 5 years' time, right?

Yes.

So I guess where is the big change to kind of make up for that maybe drop at some point? Or I mean, how is that going to -- to think got the biggest recovery sort of cycle there?

Yes, that goes back to our business strategy. So we realize that semiconductor and data center are cyclical businesses. So at any given point in time, we think though industrial, medical, semiconductor or data center will be up. And so all these businesses go through ups and downs. But because we're diversified and we play in the high end in multiple markets, we think it makes it much easier for us to thrive during the cyclical downturns of these various markets.

We've kind of proven that out over the last few years because '23 was a down year in semiconductor. We continue to increase our R&D spending and capacity spending. '24, we had some issues in some markets. And so every year, we've had the ability to continue to increase R&D spending and increase capacity spending despite downturns in certain of our markets.

Great. That brings us to time. Thank you, Steve. Thank you, Paul.

Thanks a lot.

Thank you, Shane. Thanks, everybody.

Greetings, and welcome to the Advanced Energy Third Quarter 2025 Earnings Call. [Operator Instructions] As a reminder, this conference is being recorded. It is now my pleasure to introduce Edwin Mok, Senior Vice President of Marketing and Investor Relations. Please go ahead.

Thank you, operator. Good afternoon, everyone. Welcome to the Advanced Energy Third Quarter 2025 Earnings Conference Call. With me today are Steve Kelley, our President and CEO; and Paul Oldham, our Executive Vice President and CFO. You can find today's earnings press release and presentation on our website at ir.advancedenergy.com.

Before we begin, let me remind you that today's call contains forward-looking statements that are subject to risks and uncertainties that could cause actual results to differ materially and are not guarantees of future performance. Information concerning these risks can be found in our SEC filings. All forward-looking statements are based on management's estimates as of today, November 4, 2025, and the company assumes no obligation to update them. Any targets beyond the current quarter presented today should not be interpreted as guidance.

On today's call, our financial results are presented on a non-GAAP financial basis unless otherwise specified. Excluded from our non-GAAP results are stock compensation, amortization, acquisition-related costs, facility infrastructure and other transition costs, restructuring and asset impairment charges, unrealized foreign exchange gain or loss, and the dilutive effect of our convertible note due to the higher note hedge strike price versus the initial conversion price. Detailed reconciliation between our GAAP and non-GAAP results can be found in today's press release. Please note that this quarter, we added a new reconciliation for non-GAAP income tax and tax rate.

With that, let me pass the call to our President and CEO, Steve Kelley.

Thanks, Edwin. Good afternoon, everyone, and thanks for joining the call. Third quarter revenue and earnings exceeded the high end of guidance, largely due to record data center revenue, which more than doubled year-on-year. Total company revenue increased 24% from last year, our fourth consecutive quarter of year-over-year growth. Strong revenue, solid execution and cost savings from our China factory closure pushed gross margin higher. As a result, we delivered the second best quarterly EPS performance in our history.

This year's financial performance demonstrates the value of our market diversification strategy. By selling our industry-leading power technologies into a variety of high-end markets, we are able to generate more consistent profits and cash flow. Our markets don't generally move in sync. So there's a good chance that one or more of our markets will be strong at any given point in time. This financial stability has allowed us to continuously increase our investments in new technology, products, infrastructure and production capacity. These are the key enablers of our future growth.

Sharing best-in-class technologies across our portfolio is accelerating our time to market. As an example, high-efficiency, high-density technology blocks created for data center applications have already been incorporated into several new semiconductor and industrial products. Another example is liquid cooling, a technology we perfected for plasma power applications. This know-how gives us an advantage in the data center market, which will eventually shift to liquid cooled solutions as power levels increase. Our strong balance sheet allowed us to increase capital investment this year to capture upside demand. The payback on this incremental investment will be measured in months, not years.

In addition, our new flagship factory in Thailand, where we broke ground in 2023, is now ready to start production within months of a go signal. We believe that this factory will be able to deliver more than $1 billion in incremental yearly revenue. On the new product front, we are seeing a high level of customer interest in our new technology platforms and in our ability to quickly develop custom products based on those platforms.

Now let me provide some color on each of our markets. In semiconductor, third quarter revenue was down sequentially but about flat year-over-year. Despite some near-term market choppiness, we expect 2025 to be our second best year ever in semiconductor. Looking forward, we expect demand for both leading-edge logic and memory to accelerate in the second half of 2026 moving into 2027. The leading edge is where our eVoS and eVerest technologies have taken root. So we expect to see both revenue growth and share gain as the market strengthens. Customers have validated the yield and throughput benefits of our eVoS and eVerest platforms and continue to incorporate them into next-generation equipment.

At SEMICON West, we showcased multiple versions of our eVerest platform, each tailored to a customer-specific application. In addition to our success in plasma power, we are also winning in system power applications, adapting our latest industrial power technologies to the needs of semiconductor equipment makers. Multiple wins have begun ramping to volume.

In data center computing, revenue more than doubled year-on-year and reached another record. Our technology leadership, superior execution, and accelerated capital investment have enabled this increase in profitable revenue. We expect AI-driven demand to remain robust in the coming quarters and to drive year-on-year growth in 2026. New program wins secured over the past year are beginning to go into production later this quarter with further high-volume ramps beginning in the first quarter of 2026. In addition, we are deeply engaged with our customers in the development of next-generation power solutions, including more efficient high-voltage DC power architectures. We expect most of our current design engagements to ramp to volume in 2027 with more to come in 2028.

At the recent OCP Global Summit, we unveiled several new high-power platforms geared to meet the next-generation needs of our customers. In fact, several of our partners featured our products on the show floor. In addition, we are seeing strong interest from emerging cloud and enterprise customers seeking proven, reliable, efficient, and compact solutions for their AI racks. Leveraging in-house technology blocks, we believe that we can quickly customize solutions for many of these customers.

In Industrial and Medical, revenue and backlog again grew sequentially quarter-on-quarter as customer inventories continue to normalize. In the distribution channel, resales grew sequentially and inventories declined for the sixth consecutive quarter. Looking forward, we expect steady revenue improvement in the coming quarters. In Q3, we secured important design wins in aerospace and defense and in several medical applications. We are delighted with the customer enthusiasm for our new technology platforms, such as the high-power density Evergreen series and the NeoPower line of configurable power solutions.

Acceptance of these and other new platforms will help to drive market share gains beginning next year. In addition, our opportunity funnel continues to grow, benefiting from our strong digital marketing efforts, robust distribution partnerships and a focused sales and applications team. In Telecom and Networking, revenue grew sequentially. We also expect sequential growth in the current quarter, driven by AI-related programs.

Now for a few closing thoughts. Due to our technology leadership, development speed, and operational execution, we now expect overall 2025 revenue to grow approximately 20%. Since our Analyst Day a year ago, infrastructure investments in artificial intelligence have increased substantially. These investments are driving increased demand for our differentiated data center power solutions. In addition, high-performance AI systems are stimulating investments in leading-edge logic and memory processes, which in turn will drive demand for our latest plasma power technologies.

In semiconductor, we believe the customer acceptance of our eVoS and eVerest technologies is laying the foundation for meaningful share gain. In data center, we now expect revenue to more than double in 2025 with further growth in 2026. And in I&M, we believe that our design win pipeline will drive market share gains in the coming quarters. Our manufacturing strategy and execution have enabled us to consolidate our factory footprint and meet growing demand. Looking forward, we are taking additional actions to further improve manufacturing efficiency, enable scale, and achieve our long-term gross margin goals. In addition, we now have our 500,000 square foot Thailand factory available to ramp on short notice. Finally, with a strong balance sheet, we continue to pursue acquisitions, which meet our strategic and financial goals.

Paul will now provide more detailed financial information.

Thank you, Steve, and good afternoon, everyone. Third quarter revenue of $463 million was above the high end of our guidance as investments in operational capacity and flexibility enabled us to capture higher data center demand. Gross margin improved quarter-over-quarter and exceeded our target, driven by faster-than-expected benefits from our China factory closure and lower tariff costs. Operating margin improved 220 basis points sequentially, and we delivered earnings per share of $1.74, up 78% from last year and at the highest level since 2022. In addition, we more than doubled our operating and free cash flow over last year, even as we increased capital investments to meet growing data center demand.

Now let's review our financial results in more detail. Third quarter total revenue was $463 million, up 5% sequentially and 24% year-over-year. Revenue in the semiconductor market of $197 million was about flat year-over-year, but down 6% sequentially, consistent with near-term market dynamics. Data center computing revenue was $172 million, up 113% year-over-year and up 21% quarter-over-quarter. We executed well to respond to changes in customer demand, enabling us to capture higher revenue in a dynamic supply chain environment. Industrial and Medical revenue of $71 million was down 7% from last year, but increased sequentially again, up 4% from last quarter. Encouraging data points from our distributors include 6 consecutive quarters of decreasing inventories and continued improvement in bookings, backlog and sell-through.

Telecom and Networking revenue was $24 million, up 24% from last year's low due to timing of some programs and up slightly quarter-over-quarter. Gross margin was 39.1%, up 280 basis points over last year and 100 basis points sequentially, driven by earlier-than-expected benefits of our China factory closure, better factory loading and lower near-term tariff costs. We are pleased that we again improved gross margin despite a higher mix of data center revenue and related factory ramp costs.

Operating expenses were $103 million, flat from last quarter and slightly lower than expected due to the timing of SG&A spending. OpEx as a percentage of revenue decreased 360 basis points year-over-year, demonstrating the leverage in our model. Operating income for the quarter was $78 million, with operating margins at 16.8% of sales, the highest level since 2022. Depreciation was $10 million, and our adjusted EBITDA was $87 million. Other income was down slightly at $1.7 million on higher FX costs. For Q3, our non-GAAP tax rate was 16.6% on favorable mix of earnings and benefits of amortizing R&D in the U.S. under the new tax law. Third quarter EPS was $1.74 per share compared to $1.50 in the previous quarter and $0.98 a year ago.

Turning now to the balance sheet. Total cash and cash equivalents at the end of the second quarter was $759 million with net cash of $192 million. Cash increased by $45 million quarter-over-quarter. Cash flow from continuing operations was $79 million and free cash flow was $51 million, up 124% year-over-year. Inventory turns increased slightly quarter-over-quarter at 2.8x on higher revenue. Receivables improved from 62 to 58 days and DPO was about flat at 62 days.

During the third quarter, we paid $4 million in dividends and invested $28 million in capital equipment. We expect full year 2025 capital investments to be at the high end of our range of 5% to 6% of sales and to remain elevated for the next few quarters on investments in data center capacity, infrastructure capability and our factory consolidation strategy.

Before moving on to guidance, let me provide some comments on the impact of tariffs. The environment continues to be dynamic, requiring us to implement additional actions to mitigate the impact of new tariffs. Although tariffs were lower in the third quarter on timing of recoveries, we expect tariffs to increase in the fourth quarter and continue to be in the 100 basis point range.

Turning now to guidance. We expect Q4 total revenue to increase sequentially to approximately $470 million, plus or minus $20 million. Semiconductor revenue is expected to be down slightly, consistent with customer forecasts. We expect data center computing revenue to increase modestly from the strong Q3 levels based on mix and timing of customer shipments. In Industrial and Medical, we expect sequential revenue growth over the next few quarters, paced by uncertainty in the macro environment. And in Telecom and Networking, revenue is expected to be up slightly on demand for AI-related products.

We expect gross margin in the fourth quarter to be between 39% to 40%, with benefits of cost optimization, partially offset by increased tariff costs. We continue to believe that excluding the impact of tariffs, Q4 gross margins would be at 40% or greater. We expect operating expenses to increase to approximately $107 million on R&D program-related costs and higher variable costs given the stronger full year performance. Other income should be $1.5 million to $2 million, and the tax rate is expected to be around 17%. As a result, we expect Q4 non-GAAP earnings per share to be $1.75, plus or minus $0.25.

Now for some closing comments. Our solid performance this year confirms that AE's diversification strategy is working. By leveraging our broad portfolio of power technologies and our industry-leading engineering team, we are targeting to win in the semiconductor, data center computing, and industrial and medical markets. These three growth markets inherently come with different business cycles, mitigating industry risks while enabling more consistent revenue growth, profits, and cash flow. Driven by our success in capturing AI-related demand, we are raising our 2025 total revenue growth outlook from 17% to 20%, with data center computing revenue growth increasing from up over 80% to now more than double 2024 levels.

Based on the midpoint of our Q4 guidance, we expect 2025 gross margin to expand 240 basis points and operating margins to improve by 530 basis points, highlighting the progress we've made in improving margins and driving operating leverage. Going into 2026, we are well-positioned to deliver growth in each of our targeted markets. In semiconductor, we expect our new products and leading-edge investments to drive growth as the market accelerates in the second half.

In data center computing, next-generation designs secured this year are targeted to ramp in early 2026, resulting in projected growth of 25% to 30%. I&M is expected to benefit from our design win pipeline and ongoing market recovery to continue to grow sequentially each quarter. With plans for further manufacturing efficiencies, product portfolio improvement and ongoing tariff mitigation efforts, we remain focused on delivering higher gross margins and believe that we will reach our initial goal of 40% in the near term despite the impact of tariffs and higher data center mix. Finally, with a solid balance sheet and demonstrated cash flow generation through peaks and troughs, we will continue to look for strategic acquisitions to add scope and leverage our scale.

With that, we'll take your questions. Operator?

[Operator Instructions] Our first question is from Brian Chin with Stifel.

Sorry, this first one might be multipart. So I'm curious, what constraints were you able to alleviate allowing you to more than double data center revenue growth this year? When do you plan to begin shipping product from your new Thailand facility? And do you anticipate any efficiencies in the earlier stages of that ramp? And last part of that question, will you have the bandwidth to bring up new customers alongside your four existing cloud customers?

Brian, I'll take those questions. This is Steve. The constraints that we removed in 2025 were largely capacity oriented. So we upped our CapEx spending, which allowed us to meet upside forecast from our key customers. And so that was what allowed us to hit the doubling of data center revenue year-on-year. In addition, I think we were able to gain some market share within the programs that we were engaged with at our key customers. So I think a lot of things went right for us in 2025. At the same time, we were engaged with those customers on designs for '26, and we were successful. And so we're anticipating that those ramps will start later this quarter and further ramps will commence in Q1.

I think your second question was on the new Thailand factory. So the status right now is that the factory is fully facilitized and ready to go within months of a go signal. So our intent would be to put new customers in that factory. So the second wave customers that we were talking about. And right now, we think the timing of that is in the latter part of the year. But we could go ahead and -- this is the latter part of 2027. We could go ahead and do some prequalification work and start to bring that factory up in the second half of 2026. Okay. And finally, your last question was do we have the bandwidth to bring on second wave customers, right?

Yes, the bandwidth efficiencies when you ramp.

Yes. I think that's a really important issue, and we don't want to stretch ourselves too thin. And so with the second wave customers, we are focused on solutions that use technology blocks we've already developed so that the engineering -- the incremental engineering work is not as significant -- not nearly as significant as we've experienced with our hyperscale customers. So we think we have the bandwidth to do it from an engineering standpoint, but we're going to be very careful. We're going to continue to focus on the needs of our primary hyperscale customers, first and foremost, and spinoff derivatives for the second wave customers. And we also have the ability to scale up further in our factories, whether it's Thailand or the Philippines or Mexicali.

Great. And so if it's like an open compute design, open rack design, probably it's a lot of synergies there, it sounds like, in terms of new customers. And maybe for my follow-up question, when you think about the tailwinds to AI server power content and just aggregate demand that are likely to persist next year, can you -- I know it's early, but can you put any parameters around the magnitude of growth you might anticipate in your data center business in 2026? I know this is probably not the right way to do it, but if I just annualize the fourth quarter kind of embedded data center revenue for this year, just annualizing that would probably be about 20% growth or more next year. So any sort of framework or thoughts you have around that would be helpful.

Yes, Brian, I'll take a cut at that. So I think we've talked in our prepared comments that we would expect to see 25% to 30% growth. So that means we can sustain these much higher levels and grow from here. I think this represents what we have good line of sight to at this point. Obviously, the market is dynamic. We think there's opportunities potentially to grow faster. We also don't know exactly what the mix is as new designs kick in versus the existing design sort of phase out or phase down. So we're preparing to ensure that we have the capacity to capture upside with our existing customers, as Steve just mentioned, prepare for potential second wave of customers, and we'll see how it goes. So we tried to give some direction that we feel comfortable with, and we'll look to capture upside from there.

Our next question is from Joe Quatrochi with Wells Fargo.

I was curious on the data center side. I seem like the upside this quarter was a lot driven by just kind of catching up to some of the backlog you were unable to fulfill. I guess how do we think about that contribution in 3Q and 4Q as we look into '26?

Yes. I think it's similar. As we talked about in our prepared comments, because of our good execution, because of flexibility in our factories, we were able to capture some of this higher demand and ship more this quarter. We do, as we mentioned in our prepared remarks, I think we can continue to grow from here. So we do think this is kind of a new baseline for us, where we can continue to grow. That growth will be impacted each quarter by what the mix of our customers are asking for. And frankly, that's been pretty dynamic.

I think within the current quarter, within the third quarter, and I think that was one of the things we felt good about, we saw pretty meaningful shifts in mix amongst the different products that our customers want based on the dynamic supply chains in this market today, and we're able to respond to that. So our goal is to continue to be able to do that, be responsive, capture the needs of our customers on a real-time basis and grow from here.

And then on the semiconductor side, I think 3Q came in maybe a little bit lower than we were anticipating. But are you still kind of guide -- it sounds like you're still kind of guiding for maybe on the lower end of the mid-single-digit growth for '25. Curious, your customers are -- some of your customers are kind of thinking that first half '26 is relatively flattish to second half '25. Is that kind of what we should be thinking about for your business as well?

Yes. I think I wouldn't read too much in the short-term choppiness. I think it's a normal ebb and flow of the market. But what we see moving into '26 is particularly over the past few months, we've seen more positive signals. So we're more optimistic about '26 than we were during the last earnings call. We think Q1 is going to be similar to Q4. But from Q2 onwards, we think there's potential for significant upside. And that's going to be primarily due to our new products, also helped by, I think, positive movements in the leading-edge logic and memory markets as well. So I think we're pretty well-positioned based on the new product design wins we've received as well as the surge in the leading edge next year.

Our next question is from Steve Barger with KeyBanc Capital Markets.

Steve, I think you said customers that validated yield and throughput on eVoS and eVerest are leading edge. First, just as that evolved, did you have an early adopter customer and then follow-ons? And then longer term, can you talk about what this means for both leading edge and then for memory in terms of your ability to drive revenue and take share?

Yes. Steve, I would say we had multiple early adopters. I think we first launched these products back in mid-'23 at SEMICON West and almost all the customers were interested at that point. So we've had multiple parallel efforts. And so we're pretty far down the road as far as securing these design wins. And what we've said is the conductor etch and deposition wins will go to volume first, next year. And then we anticipate the dielectric etch wins will start in '27 from a revenue standpoint. And we still are very confident in that.

So looking forward, I think those products, whether it's eVerest, eVoS, NavX or other derivatives of those products, that's what's going to drive our market share moving forward. And so we think we have a chance to really run the table with eVerest, eVoS, and NavX and win every opportunity we're competing for today. And that's going to drive meaningful market share gains for us.

And I know it depends on how those markets evolve. But just from a TAM standpoint for the new products, is it bigger on leading edge? Or is it bigger on memory potentially?

It's hard for me to say, Steve. From our perspective, what we see is our ability to gain incremental share in conductor etch, where we're the leader already, but more importantly, to gain a strong foothold in dielectric etch, where we have very little market share today. So for us, the upside is quite substantial. I'm not exactly sure how it's going to divide between logic and memory, though.

Understood. And then for my follow-up, Paul, it seems like '26 is shaping up to be a solid growth year. Mix is probably going to be pretty positive. Is it reasonable to think about incremental margin for the year in line or better with what we're going to see in 2025 as you think about flow-through and how you manage the business?

A couple of thoughts. I think, first of all, we've made a lot of progress in 2025, as you saw, over 200 basis points improvement in gross margins. Some of that's driven because we've made a big step forward in our cost down activities. I think the larger part of the 200 to 250 basis points improvement in manufacturing costs will have realized exiting Q4. And so I think that's been very positive. We won't see necessarily that big step down again.

But in terms of an incremental perspective, we'd expect to continue to see, obviously, the benefits of volume go through. We talked about that being 100 basis points for every $50 million of revenue. I'll just remind everybody that, that math, it changes as you scale, right, because you're contributing the same amount of income, but it's on a higher level of revenue. So as that goes up, there is some impact to that. I think starting at sort of this $450 million to $500 million, the same math gets you something like 50 basis points to 70 basis points for every $50 million of revenue. So we would still expect to see that flow through.

There will be some headwinds to that. Obviously, we had some tariff impact. We'll hopefully be able to mitigate that. The data center mix, frankly, has been a little bit of a headwind, but we've been encouraged that we've still improved margins despite almost doubling the data center revenue this year. We think that over time, we can continue to mitigate the impact of that mix. There will be some minor impact. But on balance, we feel very good about our ability to get to 40% still in the near term.

And our long-term goal remains the same to get to 43% as we approach that $2.5 billion organic and $3 billion inorganic number. I think if you took the tariffs out, we'd say we were on track or ahead of that target today. But the world is not static. There's things -- good news and bad news that comes all the time, and we'll continue to manage to get to our goal. We feel very good about our ability to get to the goal in the long term.

Our next question is from Krish Sankar with TD Cowen.

Paul, thanks for the color on the 25% to 30% data center growth next year. I'm kind of curious, when you look at that, it looks like next year, data center revenues are going to be kind of like closing in on your semi revenues. Just as a follow-up to the previous question, how would that impact your gross margins in 2026? And then I had a follow-up.

Yes. Obviously, that mix plays a factor. We've talked about seeing up to 50 basis points, plus or minus. I think as data center becomes a bigger factor, that could increase. But as I mentioned earlier, our goal is to work to offset that. And we've done a pretty good job of that so far. I would say, certainly, in the near term, our goal would be get margins over 40% and continue to increment them from there. But a little bit will be the timing based on our relative markets. Over time, we certainly feel really good about getting to the 43% and certainly getting above 40% in the near term.

Got it. Very helpful, Paul. And then a follow-up for Steve. I'm curious kind of where you stand on the next generation of high-voltage DC, 800 volt from NVIDIA, 400 volt from OCP. And it seems like that's probably a 2027 event. But at the same time, there are some incremental costs associated with it because of higher voltage safety certification, et cetera. So I'm kind of curious, Steve, any thoughts on that 800 volt, 400 volt, how to think about opportunity for AEIs in that scenario?

Yes. Yes, Krish, I actually indirectly mentioned this in my script, but what I said was we're fully engaged with customers on high-voltage DC solutions, which include 800 volt, by the way. So we don't talk about it that much because we prefer to keep the specifics of our development efforts confidential out of respect for our customers. But we're closely engaged with our key customers to develop reliable, efficient, and compact 800-volt solutions. These are not going to go to volume next year, but we think they will start to go to volume in '27 and '28. And I think we're very well-positioned there. The other advantage we have with these customers is that we're already engaged with them on today's generation solution as well as the solutions coming out next year. So this is kind of a natural evolution for us as we move over time.

Our next question is from James Ricchiuti with Needham & Company.

I was just going back to the Analyst Day and some of the commentary that you had about the data center market and the fact that you were targeting either being #2 or potentially sole source applications. I'm just wondering, in all of the demand that you're seeing in the market, can you give us a sense as to whether you've gained share among your leading customers?

I get that question quite a bit. And I'll tell you, we don't measure share in data center because ultimately, what we're trying to do in data center is to generate reasonable gross margins. And so at the same time, we're growing our business. We've doubled it year-on-year. So we focus on healthy gross margin. We selectively engage with customers, and we try to maximize our share in the programs we're engaged on. And so what that has resulted in is a portfolio that's generating gross margins that are just under corporate average, but a lot better than they used to be.

We're going to continue doing that moving forward.

So I think the way to think about our business is that we're creating high-value products for a limited set of customers. And that's our strategy with the hyperscalers and the first wave customers. As the second wave customers engage with us, we're trying to reuse what we've already developed. We have extensive technology blocks in the company. So I think we can handle that without reducing our engagement with our key hyperscaler customers.

And just with a follow-up question on the M&A pipeline. I'm wondering with the strength in the data center business, have your priorities changed at all? I know initially, you have been talking mainly about looking at opportunities in the I&M area of the business. But I'm just wondering if the priorities have changed at all, just given what you're seeing out in the market.

The short answer is no, the priorities haven't changed. And let me just explain why. I think we made substantial investments in the data center business on two axes. One is capital investment. And we've invested in new infrastructure. We've upgraded factories. We've invested in capacity. And then we've invested in new development centers for data center. And it's actually quite expensive because we're handling very high power, very high voltage. And so that takes a lot of money, but we've invested a lot of money in data center over the last 2 years.

The other thing we're investing in data center is people. So we've got a pretty broad network of design sites around the world that are well coordinated. And so we've been bringing in additional engineering talent, which helps us with our customers, helps keep our development speed strong. I think moving forward on M&A, we're still focused on industrial medical because that's an area where we think it's highly fragmented, and we can do a partial roll-up and create a nice third leg of the stool, the semiconductor, data center and industrial medical.

Our next question is from Scott Graham with Seaport Research Partners.

Congratulations on a good quarter and guide. My question is also about no surprise data centers. And really, in covering other industrials that serve this market, we have seen pretty much an acceleration in demand quotient each quarter in 2025. And obviously, you guys saw that you gave guidance after the first month of the quarter for the third quarter and you beat that number. So what happens if that happens again in the first half of next year, which if you're a betting person, you'd have to say you'd bet on that. And are you prepared to take your $1 billion facility here and kind of really get it going in the second half of next year to meet that demand? And if so, is there a cost to that and a margin impact?

Why don't I take the first part of that, and I'll turn it over to Paul to talk about margins. But the answer is yes, we're ready to go. I think we're conscious of the timing. I actually think it's a fairly good situation, though, because if we went to volume in Thailand with data center, then you have a pretty high-volume product to absorb our fixed cost. And so I think from a financial standpoint, it makes better sense to go to volume first with data center products than it does with semiconductor products.

Yes, I think that's right. And I'll just say that we have always contemplated our Thailand facility in our gross margin goals. So that's not a new or incremental thing. Obviously, part of that is supported by the volume. And I think on balance, if you look at where we're at as a company, we're ahead on volumes overall as a company versus we thought a year ago in our Analyst Day, and we're certainly ahead in quite a lot in data center. So we'll manage that based on, as Steve said, kind of as we see the volume materialize and the right amount of volume to utilize that factory. Our goal is to make sure that we're prepared and we can capture as we win business or as our customers ramp that we can capture that business either with our primary customers or the second wave customers as we've talked to them.

Now inevitably, I think there's some ramp-up costs that always happens. Frankly, we're seeing that today. I talked about that last quarter. We still have some of those lingering ramp costs from the faster pickup in data center in our existing factories. So there will be some of that. Hopefully, that's something that we can manage within our model. We've done that so far, and our goal would be to continue to do that. And look, as data center grows as a percentage of our portfolio, then as I mentioned, our goal would be to continue to achieve our margin goals and certainly to get above and sustain ourselves above 40% even on higher data center growth, if that materialized.

Got it. I guess my other question would be around semiconductor. So away from your eVoS and eVerest platforms, are you comfortable with your wins and product refreshes so that when WFE does kick back up, you're ready elsewhere in the portfolio?

Yes. I think your question is really about our mainstream products and our service business. And yes, I'm pretty comfortable. I think that our factory in Malaysia, where we produce most of these products is very capable, and we've maintained strong staffing, and we have excess capacity there. So when things turn up, we're ready to go. And that's in addition to what we're doing on the new products.

Our next question is from Dave Duley with Steelhead.

Congratulations on nice results. If WFE grows, let's say, in the 5% to 7% range in 2026, I'm wondering what your semi business can grow. And maybe as a follow-on to that, could you help us understand how many major wins you have ramping -- new wins you have ramping in 2026 in semi.

Yes, in semi. So we haven't really articulated a number of wins. What I've said is that we're engaged across a wide variety of customers. And there's a good chance we can run the table. We can win everything where we're competing. So I think it's very positive for us. But we haven't gone into the detail there. We're not very good at predicting WFE. But what we do know is that over time, we tend to grow faster in WFE. Now in any given quarter or year, you're going to have variation. But if you look over the past 3, 4, 5 years, we're definitely growing the market. I think we'll continue to do so.

Okay. And how many 10% customers did you have in Q3? And I guess I'm trying to understand, will one of these hyperscaler customers become a 10% customer?

Yes, Dave, we -- that's an annual disclosure. So we don't disclose that on a quarterly basis. I mean we'll see at the end of the year. But given the growth we've seen in data center, it's certainly possible we could have a data center top 10 customer.

Okay. And then a final question from me is you talked about this new factory in Thailand. You have a ton of capacity there that couldn't be used for future ramps. Does that imply that you might close other factories or do further consolidation? Or has that already been done?

Yes. That consolidation has largely been accomplished. And we mentioned the biggest move was closing our China factory, which was completed in Q2. So from this point forward, it's really a growth story. I think we've done a good job as a company, reducing the number of factories, but at the same time, increasing the capabilities of the remaining factories. So I'm very happy with where we're at. And we're continuing to expand our output in our existing factories, which are primarily in the Philippines, Malaysia, and Mexico. And then we're ready to start filling our factory in Thailand.

[Operator Instructions] Our next question is from Rob Mason from Baird.

Maybe this is a -- just to get a clarification. We talked about ramping when Thailand ramps, it would have some of the newer customers. But I thought I also caught you talking about some of the emerging customers, enterprise customers participating in '26 in data center. So I was just maybe trying to get a sense of what the -- how you think the mix in data center in '26 would look between those.

I think in '26, the mix will be heavily weighted towards our existing customers, the ones we have today that are driving our business. I think in the second half, you may see some contributions from new customers. But we think most of this is going to be weighted towards the second half, even as late as the fourth quarter. But yes, I think it would be good for the company to bring some additional customers on board in a controlled fashion, and we have the capacity to handle it.

And should we think that the margin profile there would be similar to what we're seeing right now, gross margin profile?

Yes. Yes, it would be.

Okay. Just as a follow-up, last question. The -- with OpEx stepping up here a little bit in the fourth quarter, how should -- Paul, should we think about maybe the run rate as you enter '26? It sounds like you're comfortable with where your operating leverage should be, but I'm just see if you could put a finer point on OpEx trends.

Yes. I think our model is pretty similar. We've talked about since the beginning of the year, OpEx increasing $2 million to $2.5 million per quarter. That's kind of what we've done. Now Q3 was flattish. That was mostly timing. So we kind of end up making up for that in Q4 sort of at this $107 million run rate. I think as you look into next year, you should expect that to continue kind of at that pace. It could ebb and flow a little bit. There could be some quarters where it's a little more flat, others where it's up more as we make some investments.

Our overall goal is to grow OpEx no more than 50% of revenue growth. Obviously, in '25, we've been well below that. I think we've grown OpEx like 6% on the 20% growth number. So we've kept it well in control. But you should certainly think about it continuing to increase in that sort of $2.5 million-ish per quarter as we fight inflation, we have select merit increases. We have some select investments to capture some growth opportunities.

Thank you. There are no further questions at this time. This does conclude our conference call for today. Thank you again for your participation. You may disconnect your lines at this time.

Thank you.

Welcome to Day 2 of Citi Global TMT Conference. My name is Atif Malik. I cover U.S. semiconductors, semiconductor equipment and networking equipment stocks here at Citi. It's my pleasure to welcome Steve Kelley, President and CEO; and Paul Oldham, EVP and CFO from Advanced Energy. We also have Edwin Mok, VP of Strategic Marketing and friendly neighborhood Investor Relations group in our audience. Welcome, guys.

Yes. Thanks.

Good to be here.

And before we begin, Paul has a few comments regarding Reg FD.

Great. First, thanks for having us here, Atif, and for everyone for joining us today. Just a reminder that any statements we make today may be subject to a number of risk factors, and we'd encourage you to have a look at our SEC filings for a further discussion of those risks. Also, we had our earnings release on August 5 and so we do not intend to make any comments on guidance or any updates to guidance at this meeting.

Great. Steve, it wasn't too long that you reported your Q2 results. Stock has done well this year. Can you start us off with a quick summary of the quarter.

Yes. The second quarter was good for us. We came in ahead of our guidance. And we benefited from a lot of strength in the AI data center market. So we were up almost 50% sequentially in data center. So we're off to a good start this year. We also talked about the fact that we think we can sustain this level of data center demand through the end of this year, if not exceed it. Talked about semiconductor. We had a couple of new wins for eVoS and eVerest platforms in conductor etch and deposition applications. And we also talked about industrial and medical in the call. So after 7 quarters of correction, we finally showed some growth in Q2 -- and we think that's the beginning of a trend where we're going to see sequential growth in Q3, Q4 and into '26.

Great. For the total company revenue, you mentioned 17% growth for this year. Can you break that down for each of your markets?

Yes. So we said during the call that we will grow 17% year-on-year. Now a large portion of our growth will come from data center -- so year-on-year, the data center business will grow 80%. That's being driven primarily by AI data center build-outs. But we're also seeing mid-single-digit growth for semiconductor Industrial Medical will decline for the year despite the fact we saw a turn in Q2 and telecom and networking is relatively flat.

And what drove the revision on the semiconductor revenue outlook?

Yes. So we work closely with our customers, and we just saw some reduction in their forecast for the second half this year. So nothing dramatic. I think it's important to take a step back. And if you take a look at our revenue expectation for 2025, it's our second best year ever. The best year was 2022, that was a crazy year. That was the COVID recovery year. So despite the fact we've taken a little step back for second half, 2025 is going to go down as a very good year in semiconductor for the company.

I remember a couple of years ago, you launched eVoS and eVerest products at SEMICON West. Revenue from your next-generation plasma power products is expected to double this year and reach over $10 million. Can you talk about where you see the momentum, whether it is for 2-nanometer or other leading-edge logic and memory applications.

So we launched these eVoS, eVerest and NavX products about 2 years ago. And the pull on those products from our customer base has been really impressive. So the revenue from some of those early wins in conductor etch, in particular, is starting to show up in our revenue line this half. So that's why we said this year, our new product revenue coming from those 3 products would be between $10 million and $20 million. Now we're going to see that accelerate next year. And then in '27, we'll see further acceleration as dielectric etch revenue starts to hit our revenue line as well.

So we think that these 3 products, eVoS, eVerest and NavX are really the foundation for gaining market share for the rest of this decade. And just to answer your first question, right now, they're being evaluated across all the leading-edge processes, both in logic and in memory. These are very challenging processes for our customers. Some of the structures they're trying to build are very difficult. They're trying to build very -- and trying to etch very deep holes into the silicon structure, and they need new technology. That's why there's been such a pull on these eVoS and eVerest products because they're considerable step ahead of the existing technology.

Steve, just to level set for the audience sake, if you can just talk about where you're strong, whether it's dielectric or metal side of etch and where a competitor is stronger.

Yes. So historically, Advanced Energy has been very strong in what's called conductor etch. And so we've got most of the market share there. The other type of etch called dielectric etch, we've not been strong in. So we have a very small position there. So the opportunity for these new products is for us to establish significant beachhead in dielectric and then to expand our conductor etch share. So in addition to the design wins on leading-edge processes, I think there's a second step where, in some cases, our customers may be able to use our technology to upgrade existing tools. So it's an upside opportunity, not yet confirmed, but certainly a possibility.

But it remains a duopoly type of market. It's a good market for both of you guys, right?

Yes. I would say there's 2 major players in the dielectric, DAIHEN in Japan and MKS in the U.S.

Okay. And then I know when my clients are calling me to kind of explain to them what these power supplies are and they're thinking of something like car battery or something. I'm going to tell them these are not the super complicated RF match and all this stuff. Can you talk about the competitive advantage that you have on these new products and then opportunities in dielectric and...

Yes. Maybe instead of going into too much technical detail, I could talk about the benefits of our products. Because the issue our customers have with these leading-edge processes is the existing technology results in worse yield and lower throughput. And the 2 most important metrics in a wafer fab are throughput, wafer throughput and wafer yields. And so that's why these new products that we have introduced, eVerest and eVoS and NavX. They allow the customer a greater control over how the energy is pulsed into the plasma chamber. So they could use traditional RF generation or they could use pulsed D.C. So we have both of the major mechanisms for delivering power into a plasma chamber. And we think they're both best in class.

Before we move on to data center compute, just curious in terms of your outlook into next year. If you have any visibility to understand you guys are suppliers to your OEMs, and it's difficult to see next year, but any thought of early predictions you have on semiconductor market next year.

Yes. We're bullish on the market, on the data center market, primarily because of the forecast that we're receiving from our customers. Also because of the announcements from the hyperscalers about their investment plans for next year, which are pretty robust. So we think the investment plans are forecast. The fact that power consumption continues to increase are all factors that work in our favor in data center.

So on the data center compute side, obviously, AI has been the biggest growth driver in the market. Can you talk about where you participate there? Who are you working with and your dollar content opportunity?

Yes. So this is a sensitive area because our business is really customized solutions for hyperscale applications. And so we don't talk publicly about exactly what those solutions are. But we work closely. So one of the things that's changed in the data center market is with the advent of AI data centers, the cadence of new solutions has sped up tremendously because we're seeing solutions change every year now. And that's tied pretty closely to the introduction of NVIDIA GPUs. And so what that has done is it's drawn us closer together with our customers because they need to work with us from the beginning of the design so that we can have the power solution ready when they have the rest of the system solution ready. And so we've gone very deep with a select group of customers, and we have a number of projects with each of our target customers. And the issue really, the only constraint we have is engineering bandwidth. So we have to be careful that we don't get involved in projects that aren't going to pay off. We've done a pretty good job of it.

Again, just for the sake of the audience, are there OpEx synergies or revenue synergies between the power supplies you're doing on the semiconductor side and what you're doing on the data center side?

I'll talk about the revenue and if you could talk about the OpEx.

Sure.

So I think one of our advantages as a company is that we are focused on precision power. And think about that as the high end of the power delivery market essentially. And we developed technology blocks for various applications. And we do that in a modular way. And that means we have hardware modules, software modules. That allows us to reuse technology across the company. So today, we use technology developed in hyperscale in some of our latest and greatest semiconductor products and vice versa. We have semiconductor cooling technology, liquid cooling technology that we can deploy for data center when we reach that point sometime in the next couple of years. And so this is a big advantage. We have more technologies under 1 roof than I think any of our competitors. So moving forward, that's a way to boost our efficiency as well as improve the development speed of new products.

Yes, I think those efficiencies of using these technology blocks across the enterprise saves us time and development. It also gives us arguably best-in-class capability as we're able to leverage these different parts. And in some cases, we in-source to ourselves, so we'd have to buy it from a third party for some of these building blocks. And that increases our speed to time to market as well as lowers our cost.

And then staying with you, Paul, how much incremental CapEx are you spending this year related to data center? And how much do you plan to spend for next year?

Historically, we spend somewhere between 2% and 4% of revenue in CapEx. We increased that a little bit about a year ago when we talked at our Analyst Day, and then we increased it again at the end of the first quarter as we saw this building demand for data center products. So today, we expect to spend somewhere in the neighborhood of 6% or a little higher of revenue of CapEx. And that will support our ability to improve both our capability, our infrastructure to support this much higher power products that are going through our factories as well as capacity just actual manufacturing equipment as we go through the various steps of manufacturing. We've been able to grow this business a lot. It's more than doubled in Q2 than it was a year ago and up 50% from Q1. That's been tough to stretch that much. And so I think getting this capacity in place to allow us to operate at a more normalized rate, actually lower our costs a little bit from a ramp and cost of quality perspective and give us optionality and ability to capture upside in the future.

Good. And the mix of GPU versus ASIC customers, I know it's sensitive, but the power profile is quite different for GPUs and a lot more power than ASIC. What's your mix for...

Yes. We don't have any insight to that. So the customer doesn't need to disclose their design to us. What we do is we take their spec, so many inputs, so of the outputs and then we work around that spec. So we don't know exactly the mix between NVIDIA GPUs and ASIC.

Let's talk about the industrial and the medical market. You talked about industrial and medical bottoming in Q1, and we did see some sequential improvement in continuing into second half. What's the pace and the scale of recovery? And is there any demand pull in?

Yes. So we haven't seen any demand pull in for tariffs. We ask the question every month and just really hasn't appeared on our radar screen. So I get asked the question why are we convinced that it's a recovery. A couple of reasons. One is we sell half of these industrial medical products through the distribution channel, and we track the channel metrics every month. And so for 5 quarters, the inventory has come down. We've seen resales either go up or stabilize for the last few quarters. And we see basically the -- just the inventory is getting pretty close to quarterly demand. So we think in that channel, we're close to a healthy balance, a return to equilibrium.

The other data points are anecdotal. But in my discussions with industrial medical customers, more of them are telling me they've worked their way through the inventory issues because a lot of industrial medical customers get stuck with a lot of inventory at the end of the COVID supply chain crisis. And it's taken them the better part of 2 years to work their way through it. So that's good news. It means the demand for our new products will be returning. The third data point for me is the amount of design wins. We've earned Industrial and Medical over the past 3 years. Now not all of those go to production, but now that some of the inventory is getting cleared out, many of those will go to production. So I think we will start to gain market share in industrial medical next year.

And what's the underlying driver for replacement for your systems. The data center is clear, it's the Rubin and the next platform that's going to drive the powerful profile. But is it more of the replacement cycle that drives the steady demand in this market?

Yes. So Industrial and Medical is an interesting market because it's very sticky and the life cycle is typically very long. So once you're in, you're in. Some of these applications continue to buy customized power supplies for 10, 15, 20 years. So it's difficult to displace incumbents. That's great if you're an incumbent, not so great if you're trying to displace somebody else. So our wins are on new products, new applications, new products our customers have developed. So in addition to the organic growth, to speed things up in industrial medical, we would intend to make acquisitions in that area to build critical mass.

All right. And which end markets are stronger within I&M and which are weaker?

Yes, we're seeing strength, obviously, in MilAero. That area is booming. We're seeing strength in test and measurement, in factory automation. Those are the 3 that spike on right now.

Right. Let me pause there and see if there any questions in the audience. If you have a question, please raise your hand.

I just thought if we could talk a little bit more about the semi equipment business. You talked about new products that start to really help a little bit this year, much more next year or year after. But can you talk about your broader outlook for semis for next year? How do you think about that market puts and takes?

Yes. What we said about the semi market for next year is we expect a flattish market. And if you dive a little deeper, we expect to see continued weakness for trailing-edge processes and that means China as well as the Western trailing-edge demand, offset by leading-edge demand, and that's going to be driven by continued demand from AI applications.

Okay. In data center, can you talk about like hyperscalers? Can you talk about your engagement with enterprise customers because there is like there is talk about broadening of AI.

Yes. So our primary focus is on hyperscalers because that's where the volume is. But I mentioned in the earnings call a few weeks ago that we're also seeing interest from other customers. which would include enterprise customers. And the way we're addressing those opportunities is to see how much existing technology we can reuse. So if we can do something that doesn't drain engineering resource to any significant degree, then we go ahead and do it. But we're not going to deprioritize the hyperscale requirements to go chase enterprise opportunities. So it's a fine balance.

Just wanted to hear your comments on kind of ability to scale in the data center. So I think you alluded to the fact that you're kind of gated by engineering capacity, and therefore, you want to prioritize the right places. But let's say that this growth continues, how confident do you feel in your ability to meet demand? And then sustainably over a long period of time, get new talent into the market to serve what's a pretty big opportunity.

Yes, it's a good question. So there's 2 elements to that. One is actual production capacity, and we're making substantial investments. We'll continue to make substantial investments to just stay ahead of our customers on their demand. The second is human capacity. It's development capacity. So at AE, we have roughly 1,400 design and development engineers. We have a lot of people and a significant number of those are working on data center applications. And we have the ability to shift talent around, which we do. And so we've moved more people into that group.

But at the end of the day, the other thing we are doing is expanding our footprint from a development standpoint. So most of our engineers are located in Asia, in Hong Kong, Taiwan, Philippines, but also in the U.S., in Minneapolis on the West Coast, we've established a center on the East Coast, in the Massachusetts area, where we've added 50 engineers in the past 2 years. So we're moving to where the engineers are. But I think there's always a limit on how much you could do. And the key for us moving forward is making sure that we get the best return on investment for engineering team.

Paul, on the model, the tariff impact to 2Q gross margins were around 100 basis points. Given the increased tariff rates in the Philippines and Malaysia, do you expect greater impact in second half? And what's kind of baked into your guidance?

Yes, it's a good question. Obviously, tariff rates increased from roughly 10% globally. The higher most Southeast Asian countries are now around 19%. And including in Malaysia and in the Philippines where our factories are. At the same time, we've been able to mitigate some of the costs we incurred in Q1 and roughly, we think those kind of offset each other, the increased impact of tariffs versus actions we've been able to take so far to mitigate. So we expect tariffs to continue at -- we said at this rate, a little higher. That's embedded in our guidance. So even with that tariff headwind, we expect gross margins in the third quarter to improve a little bit to sort of mid 38s. And we expect the fourth quarter to be between 39% and 40%. That improvement is largely driven by the closure of our factory in China and continued consolidation of our manufacturing footprint.

Now if you exclude tariffs, we would be running over 40% by the time we exit the year, which was our goal. So we think we're -- in terms of what we can control, we think we're on track or a little bit ahead in terms of improving gross margins in the near term. As we look out beyond the near term, we think there's other levers that we have that can offset the impact of tariffs. Those include continued optimization and efficiency improvements in manufacturing, driving material costs, lower -- and obviously, as our new products start to come in and take a bigger impact on mix, those things can all help us overcome the tariff impact, assuming it sort of stays in this range.

And then with the closure of the China factory, you guys have been on this kind of consolidation sort of time line for the last couple of years. Can you just help us understand where you are, how much more consolidation is needed across your manufacturing sites?

Yes. So this last quarter in June, we finished production in our last China manufacturing site. That means we've closed 3 sites in China in the last 4 years. So that effort was obviously put in place a while ago to manage geopolitical risk. And of course, it turns out that, that seems to be the right thing to have done. In addition, our customers wanted us for the most part out of China. So we're fortunate that that's been done.

In parallel, we've been on the path to do a rooftop consolidation and to consolidate from 15 factories essentially down to about 5 factories. We started with the biggest ones first where they have the biggest financial impact, and we're well through that process. As we go into '26, there is still a number of smaller ones that we'll have to a little bit of cleanup on. But we're excited because this gives us a much more streamlined footprint, gives the ability to improve efficiency improve quality, common processes across our factories.

Okay. And how does success in the data center business impact your gross margins? And what's kind of baked in that 2030 targets that you shared at the 2024 Analyst Day.

Yes, it's a good question. If you look at our margin profile, we've been pretty open that our data center margins are below the corporate average. The good news is we've focused on our strategy to play more in the customized part of this market, the higher power that plays to our strengths, we've been able to close that gap. And today, the gross margins we've said, are approaching or near corporate average. So they're not nearly as much of a drag as they used to be. In fact, the way we think of it is changing the mix could have a plus or minus 50 basis point impact.

I think a proof point of that is in the first quarter, we had a pretty heavy mix of semi products. In the second quarter, that shifted, and we had a much heavier proportion of data center products. In fact, I think it was close to 1/3 our highest ever, and yet margins actually improved a little bit sequentially. So we've been able to weather this change in mix because we've been able to make pretty significant improvements in the margin -- underlying margins of those products.

All right. And on the capital allocation, what are the priorities? And what's the strategy on M&A?

Maybe I'll cover the priorities and Steve can talk about our approach to M&A. But clearly, we think the best use of capital is to continue to grow the company. Obviously, we're investing some of that into growing the company. Our CapEx, as I mentioned, is going to be higher for the next several quarters, and we'll continue to invest in engineering and other things. But we think we can grow the company through smart M&A where we can add essentially scope to what we sell and then leverage the scale of the company. that will consume, we think, the majority of our capital. We also have an opportunistic share repurchase plan in place. So from time to time, as we see dislocations in the market, we're able to be pretty aggressive on that plan.

In fact, in the first quarter or in the second -- early in the second quarter, we actually repurchased over $20 million of stock over the course of a couple of weeks to take advantage of the market situation we saw. We bought that stock back at a little over $83 per share. So in hindsight, that's a pretty good investment, and we're able to get quite a few shares back for that.

Yes. So our M&A strategy, we think we can grow pretty well organically in semiconductor and in data center. We think we've got the ingredients in-house. But the area where we need to bulk up is in industrial medical. And I mentioned earlier in this presentation, that we think there's a partial roll-up strategy that will help build critical mass in industrial medical and accelerate our share gain program. So that's where the primary focus is. The secondary focus is on technology tuck-ins like we did last year with a company called Airity. And so that was a big success for the company because we were able to deploy the RD technology immediately into some of our more advanced platforms. So we're actively looking for acquisition opportunities.

All right. And then end market, we didn't touch on was the telecom and networking where we have seen a recovery in terms of spending from the service providers. And I was curious if you have also participated in that recovery and what's the outlook?

It's an interesting market for us. it's just not big enough for us to dedicate a lot of design resource. So again, we shifted over to hyperscale and to a lesser extent, Industrial and Medical. So we're treating the telecom and networking market opportunistically, which means if we could put some technology blocks together to satisfy the customer need, then we'll do it, but we're not going to peel off engineering teams away from hyperscale or Industrial and Medical.

Right. We have a few minutes left. If anyone has got questions, you can come up and talk to the management team. But thank you, Steve and Paul coming to the conference.

Thanks, Atif.

Thank you, Atif.

Thanks, everyone.