Thank you for joining us. I am Greg, your conference operator for today. I would like to welcome everyone to Bloom Energy's Second Quarter 2025 Financial Results Call. Now, I will hand the call over to Michael Tierney, Head of Investor Relations. Michael?

Thank you, and good afternoon, everybody. Thank you for joining us for Bloom Energy's Second Quarter 2025 Earnings Call. To supplement this conference call, we furnished our second quarter 2025 earnings press release with the SEC on Form 8-K and have posted it along with supplemental financial information that we will reference throughout this call to our Investor Relations website. During this conference call, both in our prepared remarks and in answers to your questions, we may make forward-looking statements that represent our expectations regarding future events and our future financial performance. These include statements about the company's business results, products, new markets, strategy, financial position, liquidity and full year outlook for 2025. These statements are predictions based upon our expectations, estimates and assumptions. However, as these statements deal with future events, they are subject to numerous known and unknown risks and uncertainties as discussed in detail in our documents filed with the SEC, including our most recently filed Forms 10-K and 10-Q. We assume no obligation to revise any forward-looking statement made on today's call. During this call and in our second quarter 2025 earnings press release, we refer to GAAP and non-GAAP financial measures. The non-GAAP financial measures are not prepared in accordance with U.S. generally accepted accounting principles and are in addition to and not a substitute for or superior to measures of financial performance prepared in accordance with GAAP. A reconciliation between the GAAP and non-GAAP financial measures is included in our second quarter 2025 earnings press release available on our Investor Relations website. Joining me on the call today are K.R. Sridhar, Founder, Chairman and Chief Executive Officer; and Maciej Kurzymski, our Acting Principal Financial Officer. K.R. will begin with an overview of our progress, and then Maciej will review financial highlights for the quarter. After our prepared remarks, we will have time to take your questions. I will now turn the call over to K.R.

Good afternoon, and thank you for joining us today. Bloom had an excellent quarter, the highest revenue and most profitable second quarter in our 24-year history. When the company was founded and again, in our IPO prospectus 7 years ago, we painted a bold vision to become the power provider of choice for the digital world. Over the last couple of calls, I've told you that our business is at an inflection point as demand for clean, reliable and rapidly deployable power is surging. Now there is tangible evidence. Six months ago, we announced a strategic partnership with a major U.S. utility company, American Electric Power. Yesterday, AEP announced that Amazon Web Services and Coralogix, both data center operators, are deploying Bloom systems in Ohio. AEP's CEO, Bill Fehrman, noted that demand for power is, 'Growing at a pace I haven't seen in my 45-year career.' But he notes interconnection agreements take 5 to 7 years in many U.S. states, even for AEP, the largest owner of electric transmission systems in the U.S. To avoid such a long delay, Bill also added that AEP is giving its customers solutions so they can come online quicker. Fuel cells will get AWS and Coralogix up and running quickly. Indeed, AI companies need power at AI speed, waiting 5 to 7 years is untenable and Bloom moves at AI speed. Just last week, for instance, we announced our partnership with Oracle to power their AI data centers. We have committed to having power available to their first data center in 90 days. Time to power is one of many value propositions Bloom brings. We are also cleaner, more reliable and more cost-effective than alternatives. Because our power systems are designed and purpose-built for data centers and other mission-critical applications, our installations do not require the band-aids that turbines and engines need to power data centers. For example, we don't need multiple AC to DC converters or specialized equipment to suppress harmonics. Eliminating these band-aids enables data centers to lower costs, increase reliability and reduce carbon footprint. We are excited to collaborate directly with Oracle to help them leverage all of the benefits the Bloom platform provides. The result, Oracle can optimize the watts-to-flops ratio, resulting in increased revenue growth and margins. Commercial and industrial customers are also increasingly valuing the velocity with which we operate. Quanta Computer, for instance, builds the AI servers that are used in the AI data centers. Their demand growth is highly correlated with AI data center demand growth. We informed you about our rapid deployment at their Fremont facility last year. Happy with our execution, they ordered an islanded load following microgrid, which we installed in Q2. We expect new orders from other AI hardware ecosystem players soon, complementing demand we see from our more traditional commercial and industrial customers. Bloom is clearly delivering for customers. And our strong fundamentals mean we are also delivering for you, our investors. This quarter, we had record profits and operating margin, the third quarter in a row that we are hitting similar marks. Our service business has been profitable for 6 quarters in a row. And for the first time ever, we had double-digit percentage margins, evidence of our increased reliability. And we refinanced our debt notes that were previously due in 2025, providing increased optionality to finance growth. One other highlight for the quarter, U.S. lawmakers and the administration restored tax credit benefits to companies who install our fuel cell systems. Our country's leaders recognize that baseload power is critical to winning the AI race, reshoring factories, creating jobs and growing the economy. The tax credit will be another tailwind as we continue to grow our business. Let me close by reflecting on our position. Bloom is in a strong place. In the 12 years since we began shipping product, we have generated over 40 terawatt hours of electricity. We have deployed more than 22,000 energy servers, our power generators, totaling well over 1 million fuel cell stacks. Each of those 1 million-plus fuel cell stacks has a unique digital twin. And over our history, we have collected over 4.5 trillion data points from the field. Now thanks to AI, we are unlocking new ways to improve our performance, reduce costs and deliver more value to our customers. We are operating at scale and are scaling with purpose. Now our robust product has robust demand. We will double our factory capacity from 1 gigawatt a year now to 2 gigawatts a year by the end of next year. Our mission has never felt more urgent, and we are ready. I'll turn it over to Maciej now, and I look forward to answering your questions.

Thank you, K.R., and good afternoon, everyone. As K.R. mentioned, selection of our fuel cell energy service by Oracle to power their cloud computing is another proof point for how well our technology is suited for on-site, highly reliable and variable load following power. I am thrilled to see further adoption of our fuel cell technology by leaders in the AI space. Beyond that, I'm going to limit my comments to our Q2 financial performance. A consistent theme at Bloom has been a relentless focus on our product cost reduction and discipline around all other spend to drive profitable growth. The first half of fiscal 2025 is evidence of those efforts. Equally important, commercial execution was strong. All of this yielded a strong second quarter and first half for us. Highlights included record second quarter revenue and gross margin and our sixth consecutive quarter of profitability in our service business. As a reminder, I will focus my discussion on non-GAAP adjusted cost and profitability metrics. For a reconciliation of GAAP to non-GAAP, please see our press release and the supplemental deck on our website. Revenue for the quarter was $401 million, up 19.5% year-over-year. Gross margin was 28.2%, 650 basis points higher than the 21.8% gross margin in Q2 of 2024, attributable to mix and level-loaded manufacturing. As we said last quarter, we took advantage of our balance sheet and our visibility into customer demand to maximize efficiency and level load our factory during the first half of the year. We expect to work down this inventory as our shipment of products accelerate in the second half of fiscal 2025. Our operating income was $28.6 million versus a $3.2 million loss in Q2 last year. Adjusted EBITDA was $41.2 million versus $10.2 million in Q2 of 2024, while EPS was a positive $0.10 versus a loss of $0.06 a year ago. Again, these are all non-GAAP results. With this quarter, we have now had a profitable service business for 6 consecutive quarters. We expect this trend to continue together with margin improvement. Finally, during the second quarter, we refinanced $113 million of our convertible note that was due in August 2025 to provide more optionality to fund future growth. It was exchanged into our existing 2029 convertible notes. Turning to the full year. We are reiterating our 2025 guidance. As a reminder, we expect 2025 revenue of $1.65 billion to $1.85 billion, non-GAAP gross margin of approximately 29% and non-GAAP operating income of $135 million to $165 million. We expect positive cash flow from operations around the same level that you saw in fiscal 2024. We also expect CapEx to be around the same level as fiscal 2024. As we have mentioned before, we expect to see similar revenue seasonality with roughly a 40-60 first half, second half split. We are committed to maintaining strong fiscal discipline as we continue to scale. To conclude, we delivered record Q2 financial results and are reiterating our 2025 guidance. Our fuel cell solution was built for this moment when on-site, scalable, reliable, low following power is required in a matter of months. We are well-positioned to meet the moment. Operator, we are now happy to take questions.

And it looks like our first question comes from David Arcaro with Morgan Stanley.

I'm wondering if you could elaborate on your recent success with hyperscalers. How are you seeing Bloom servers being used? Are they the exclusive power source for these AI data centers? Are these large-scale deployments? And maybe broadly with that customer set, could you see this Oracle partnership potentially act as an accelerant and spur more additional deals?

David, welcome to our team of analysts. It's great to have you on board. Regarding the Oracle deal, this marks our company's first direct engagement with a hyperscaler as a customer. This project involves a single AI data center, and we are collaborating with them on multiple initiatives. Notably, it will operate as an islanded power system, meaning it will not connect to the grid. We are responsible for both the primary and secondary load for this customer. This is a significant development that aligns perfectly with the architecture we have designed. They will benefit from the extensive capabilities that Bloom provides, presenting an opportunity for us to enhance the features that Bloom contributes to the data center, ultimately optimizing both capital and operating costs for the customer. We view this as extremely important, as we are the primary energy source, enabling load following. This project will demonstrate our capability to manage load at a large scale and operate efficiently at AI speeds. Additionally, it will validate our ability to deliver installations within the 90-day timeframe we previously stated.

Excellent. That is certainly a significant achievement. Following up on that, what gives you the confidence to double your production capacity? Do you have increased visibility, and does the backlog contribute to your confidence? Or are you relying on the strong underlying fundamentals of the industry indicating robust demand?

Thank you, David. Yes, we have always stated when we've been asked that we will never get past our headlights. So we have told you in the past that in the last 2 quarters, we have seen strong commercial activity. We have told you that it is very diverse, and it is high quality. At this point in time, when we look at that pipeline, it has gotten us to a level of confidence where we absolutely feel like this is the right thing to do. That's why we are expanding the capacity, number one, right? Number two, this should be fairly simple, and it should be mind-boggling for all of us, and we shouldn't get numb to this fact. The large hyperscalers put together are going to spend more than $1 billion a day on CapEx, weekday and weekend. It's more than $500 billion are going to be spent just in this calendar year by those people. So you take that number of $500 billion and you say, an order of magnitude down, at least $50 billion of power capital equipment needs to be spent to electrify that additional demand that's going to come on. And you take that and you do a simple math and say more than one sizable nuclear power plant's worth of baseload is what is needed every month. And we all know what can be done by the existing legacy electric infrastructure in this country. It cannot move at AI speed. We are an obvious solution. It is self-evident to me that our demand is going to be high. And it would be wrong for us not to go and invest that money because we have that level of confidence, not just from what we are seeing in the pipeline, but when we look at this kind of investment, and we look at where is the tail for this. This is a secular trend. It is not a 1-year or a 2-year trend. You put that together, absolutely, we should be building these factories. And the 2 gigawatt is a start to multi-gigawatts we will keep building over time, absolute confidence.

This is Michael on for Mark. I guess just a follow-up on that last question. How long do you expect it to take to build out this capacity? And I guess, what's the timeline in terms of when you expect to exceed the current 1 gigawatt of capacity?

I can't provide specific timelines right now. However, I want to emphasize that we are committed to operating at AI speed. We will ensure that we have the capacity throughout this year and the next to meet timelines that may exceed our competitors or satisfy our customers. The capacity we have established and the speed at which we implement it are built for this moment. It is not feasible to increase the capacity of any traditional factory at our rate without careful planning of the supply chain, manufacturing, and equipment. We are able to adjust quickly and scale up capacity within months, often faster than it takes for a data center to establish its own facility. This is our current position and will continue to be the case through 2026, which is why we are pursuing this expansion strategy.

Okay. Great. Maybe if I could just throw one more in. Is there any estimate on how much this will cost? Or how should we expect you to fund the expansion?

We are well-funded for our plans to reach 2 gigawatts. To give you a rough estimate, think around $100 million, which will be distributed over several quarters. We have sufficient funds to support this expansion.

Congratulations on a strong quarter. I wanted to focus a little bit on the improvement we are seeing in the operating margin. I mean, you are at already about 7.1%. And if you look at your guide for this year, it's close to like 8.5%, I think the midpoint of it. So making excellent progress on the operating margin. So help us understand what's driving it. And then when we look at 2026 or '27, would the target be to get to double-digit operating margin levels also?

Operating margin is indeed a great topic, Manav. You're observing impressive fiscal discipline. Our finance team collaborates effectively with our organization to manage expenditures, whether in operational expenses or procurement. Additionally, our commercial team excels at ensuring our customers receive great value while we also gain from the services we provide, including time to power and other factors. With this synergy and our ongoing cost reduction efforts, you can expect our operating income to continue improving over time. It will fluctuate each quarter based on mix and volume, but this year, we have strategically chosen to maintain a balanced load, which is evident in our cash and inventory. This decision will yield overall benefits for the year through our balanced factory operations.

Perfect, sir. We were on the AEP call yesterday, and they said some very nice things about you and your scalability of your product. I'm just trying to understand, you can scale up that order. I think you're doing a couple of data centers through AEP. And as you work with them, do you continue to see more AEP, BE collaboration and more deployment of your product through AEP?

As you know, the service agreement was for 1 gigawatt. And if you just look at the AEP numbers and look at where their demand and where that gap is, we are hopeful that working together we can fulfill that gigawatt and think about future agreements like this very soon. And we are working together. It's not just a hope. There is a pipeline, and that pipeline for AEP is again robust, and we work with them. And you will hear more, I'm sure, in the coming months.

Yes. I would only like to say one thing conclusively. Two quarters ago, I think I asked you a question that this DeepSeek is happening and some data center providers are pulling back and you were very firm, this is a blip, do not look at it. Things will correct very quickly. And you had great vision and foresight. Absolutely, it was a blip. Nobody is even talking about it now. So thank you.

And it looks like our next question comes from Chris Dendrinos with RBC Capital Markets.

Yes. I wanted to ask about the value proposition you all bring. And I guess maybe specifically on the combined heat power solution. I know in the past, you've kind of spoken about it and it seems really appealing from an efficiency standpoint. So I'm curious, are you deploying that with Oracle here? And if not, I mean, what's kind of the interest level? And where are you at kind of in that solution process with other potential customers?

That's a very good question, Chris. Here's what we see. In our pipeline, the interest from customers is very high as soon as they learn that we can offer the CHP solution commercially. Most customers we are currently speaking with are asking for a quick time to power. They want to know if we can come back in a few months to retrofit the combined heat and power solution. The advantage of our solution is that we can provide that flexibility, much like adding an app on your phone. We can integrate the CHP as an app, which is very attractive to them. This interest is coming not only from data center clients but also from our commercial and industrial customers who require air conditioning or steam for their factories. We are seeing this demand across the board. As you mentioned, from a value proposition standpoint, it effectively means not needing 20% of your power in a data center. It's akin to not incurring costs for power when you manage cooling through our waste heat instead of utilizing additional electricity. That's significant.

And our next question comes from Maheep Mandloi with Mizuho.

This is actually David Benjamin on for Maheep. So I understand there's a robust demand and it looks like based on prior guidance, 40% in the first half. It looks like you guys are on track towards the top end of the guidance. I was wondering just the first part is like what would it take to raise the bottom end? And then secondly, are there any concerns with pushouts from the fourth quarter due to potentially ITC being available next year?

Great question. It's actually a two-part question, so let me address the second part first, David. On ITC, I should have been clearer in my prepared remarks, so let me explain this more clearly. Our customers won't experience any gap in timing regarding ITC. They benefited from ITC last year and will continue to do so throughout this year, with these benefits extending from 2026 to 2032. There is absolutely no gap. There seems to be some misunderstanding among those reviewing this information. You're right that the BBB reinstated ITC for fuel cells starting in January 2026. However, we have secured enough volume under safe harbor, meaning our customers in 2025 won't have to wait until 2026 to access those credits. Therefore, there shouldn't be any disadvantage for customers who decide to buy or install their systems in 2025. Is that clear?

Yes. Crystal.

Okay. Now in terms of the guidance and where we are, we have reiterated the guidance. Where we fall in that range, why do we give you a range? It has nothing to do with our ability to ship the systems. It's got everything to do with if the customer is ready. Many of these things are greenfield. They need to finish their factory or their data center on time. They need to be able to connect. Gas needs to be there and like permits need to be there. So it could easily move out a couple of weeks on either side or a month on either side. None of these projects are in jeopardy of coming in or not. But when we recognize revenue will depend on those things. And until we have clarity on that, we have to give you a range.

And our next question comes from the line of Chris Senyek with Wolfe Research.

So congrats on the Oracle deal. I was just curious, since that order was expected to be delivered within 90 days, there was no change in the full year guidance. Should we assume this is already embedded? Or would you need to announce additional deals in order to achieve those targets?

We have mentioned from the start of the year and will keep emphasizing that part of our revenue will come from deals that we book, build, ship, and recognize within the same year. This is a positive development. I recall that after our IPO, it typically took 18 to 24 months to finalize a deal. The current speed of business allows us to significantly shorten that timeframe. We would be worried about this reduced cycle if we believed it was just a temporary trend, but we don't see it that way. The ongoing trend shows that the speed at which we can convert a deal into a booking is accelerating, which is excellent from our perspective.

And our next question comes from Dushyant Ailani with Jefferies.

Just the one on the opportunities outside the U.S. I think you've also mentioned Taiwan in the past. Could you briefly update us on how those conversations are progressing and what opportunities you're seeing outside the U.S.?

Look, today, when we look at our business in general, roughly 30% of that comes from international, 70% comes from domestic, okay? And we expect to continue that ratio at least through the next year or so because as all of you know and follow from what's going on, there's tremendous action here in the U.S. market, okay? And so we see that. But we truly believe in the diversity of the market. Not only are we continuing to show strength in Korea and keep that business going. We are developing new markets. I think we have mentioned to you Taiwan, Germany, Italy and the U.K. are the obvious next places we are looking at. And in all those places, we are making progress, establishing into a new market and being able to get the policymakers, the regulators lined up with something new to them, very similar to it was new to people in California and like New England when we started in the early days. That process is going, but I'm very happy with the progress we are making in those places.

Our next question comes from Colin Rusch with Oppenheimer & Company.

As you look out at the landscape of opportunities, are there situations where you could end up being used as temporary power for a couple of years and then have those servers move on to other locations? And are you starting to see any sort of incremental demand growth for longer cycle industrial, potentially chemical plants and other things that are looking to ramp and have power shortages as well as the data center opportunity you guys talked about?

Colin, that's a very good question. The answer is absolutely yes. If you recall, we completely changed how we install our systems from pouring concrete to using a skid. Think of it as grid to go. We can transport these units on a truck and easily relocate the skid as needed. It requires just three connections: gas, electric, and wireless communications. It's that straightforward. Additionally, it's even better than relying on temporary power from combustion engines because you can't easily relocate power from one combustion engine setup to another if you need to distribute power differently. With our Bloom systems, their modular nature allows you to move units between locations or adjust power distribution as desired. This is certainly an appealing option we're able to provide. Hyperscalers are particularly interested because if they secure power at one site, they can effortlessly transfer it to another site where it's needed. This characteristic makes our offering very attractive. In fact, it's an impressive selling point for us. As for concerns about needing power two years from now, anyone who believes that there will be a miraculous solution in two years is mistaken; that’s just moving from one challenge to another, not resolving the issue.

And our next question comes from the line of Sherif Elmaghrabi with BTIG.

Another tax credit question. Safe harbor aside, the BBB does give more visibility. And my understanding is that domestic content bonuses have actually been raised. So my question is, with the tax credit picture is set, hopefully set, does that put you in a position to push pricing?

Yes. So what happens with the BBB is there is no domestic content adder when it comes to the ITC, right? So I think I'm answering your question. If not, let me know, and I'll answer it. So it is a flat 30%, okay? Whereas in the previous version of the bill that ended last year, but safe harbored now, our customers can avail of either 40% or 50%, depending on whether they are not in an energy community or in an energy community. If they're not in an energy community, it's 40%. If they're in an energy community, it's 50%. So from that perspective, yes, their subsidies go down a little bit. But given how high the price of electricity has gone up, at 30%, our attractiveness will be extremely high. And you also know that every year, we bring down cost reductions between those 2 things, we don't see any issues with relation to maintaining our margins, if that's your question.

Our next question comes from Noel Parks with Tuohy Brothers.

I wonder if you could talk a bit about product development and maybe update us on your upcoming generation of the Bloom Energy server and what incremental benefits you're anticipating from that? And any comment you have on the effect on the economics for the product for you would be?

Thank you for the question. What's really interesting about the Bloom platform moving forward is that the improvements we are discussing occur continuously, quarter after quarter, with new ideas being refined, fully manufacturable, and entering production smoothly. Instead of thinking in terms of transitioning from one product generation to another, consider our product as continuously improving. We are able to integrate these advancements seamlessly. A lot of developments are ongoing, and as mentioned in our earnings script, we are maintaining our margin guidance despite a potential 4% tariff on our materials. This stability is due to the ongoing enhancements in our product. We no longer refer to our product in generational terms but rather see it getting progressively better over time. This is made possible by the real-time feedback we receive from our digital twins, which provide us with 4 trillion data points to analyze, learn from, and improve. This capability is unique in the power industry, and we will continue to see improvements rather than discrete jumps in advancement. Additionally, we are introducing new features, such as combined heat and power (CHP), load following, and the ability to operate in island mode. These attributes enhance the product further.

And our next question comes from Chris Dendrinos with RBC Capital Markets.

I wanted to return to the value proposition, and I think you touched on this in Colin's question. Could you help us better understand where you see your solution as being the most optimal compared to a natural gas turbine? It seems to me that your solution should always be preferred, but there are certainly reasons that drive some customers to choose a gas turbine. I'm trying to grasp what those driving forces might be.

Chris, I completely agree with you. It's true that purchasing IBM carries little risk. Looking at the economics, there's significant demand growth. Just six months ago, we noted that 20% of data centers might operate independently from the grid, and that number has now risen to about 40%. Data centers are increasingly indicating in surveys their interest in independent power sources. When it comes to traditional turbines, they require additional units for maintenance, meaning one operates while the other is serviced. In contrast, our Bloom architecture resembles LEGO blocks, allowing for grid-level availability at comparable rates to turbines. Our capital expenditures are at least on par, if not more favorable, while turbines consume 15 to 20 percent more fuel. This positions us for significant operational cost advantages. Additionally, our solution doesn't produce air pollution, and obtaining permits for turbines in populated areas can be quite challenging. All these factors make us easier to permit and quicker to deploy. Since time to power is crucial in this industry, our lower operating costs and competitive capital expenses mean that we stack up well against other methods of electricity generation.

And our next question comes from Skye Landon with Rothschild & Co. Redburn.

More of a clarification from myself. On the AEP announcement recently regarding the Amazon Web Services and the other company, are you able to confirm whether those projects are included within the 100 megawatts or if those are additional on top of the 100 megawatts?

It is part of the 100 megawatts that was in the PO and the other 900 megawatts are what we are working actively on the pipeline.

And our final question today comes from the line of Dimple Gosai with Bank of America.

I actually have 2, if that's okay. The first one is just trying to get more clarity or understanding around the Oracle deal. Is this a done deal? Is there a framework to deploy? Any KPIs that need to be met? Any color would be super helpful for us. And the second question is if you can talk a little bit about your capital needs or liquidity in terms of funding the manufacturing expansion.

Regarding the first question about the Oracle deal, we do not discuss specifics such as letters of intent and memorandums of understanding during our earnings calls. This is a purchase order we are currently working on, and the power will be available in a 90-day period. As for the capacity and budget required, if you joined us late, I want to avoid prolonging the call. We have already addressed this question, and it will be included in the transcript. Thank you very much. And with that last question, let me say that, look, AI is moving faster than any technology in history, faster than the Internet. And of course, orders of magnitude faster than the electricity industry. And it's demanding more, more power, more data centers, more urgency. Our company Bloom, our entire company was built for such a moment. We are ready, and we can move at AI speed. The advantages of our purpose-built digital platform has never been more obvious and relevant to everybody. And the opportunity in front of us is both massive and secular. We are confident in our strategy. We are confident in our execution. And internally, we ourselves have shifted to a higher gear because we are going to blaze forward. We thank you for your confidence in us and your continued support. We wish you a good day.