Boost Run Inc. Class A Common Stock 2026 Q2 Earnings Call

Good morning, everyone. Thank you for joining us. This is our first earnings call as a public company. I am not going to spend time on how we got here. I want to talk about where we are going. This team built Boost Run from the ground up, and we have never been more excited about what is in front of us. I will cover the results briefly, spend most of my time on how we run this business, and Erik will then take you through the financials. To set the context of today's discussion, Boost Run today is roughly a $1.6 billion company, yet we sit in the same conversations, compete for the same customers, and participate in the same NVIDIA programs as companies 10 times our size. This access is earned. We believe that if we continue to execute, Boost Run has significant opportunity to grow from here.

We believe Boost Run will play a major role in the AI revolution, and what we hear from customers and partners every day gives us that further conviction. Partners choose Boost Run because we bring capacity online faster than the market expects in accordance with reference architecture at compelling economics, and with a compliance posture that regulated industries like financial services and healthcare demand. Our pipeline today is the largest in our history in project size, number of opportunities, and diversity and quality of customer. Quarterly update. Q2 revenue was $31.1 million, up approximately 270% year-over-year and approximately 260% from the first quarter. We signed significant contracts worth approximately $1 billion in TCV in Q2.

Long-term contracted revenue, or TCV, stands at $1.9 billion, with an average duration of approximately 3 years and an average prepayment of 22%. We obtain prepayment on every deal we close. We expect to continue deploying capacity through fiscal year and into Q1 2027, bringing the full $1.9 billion of TCV live in production. We operate 6 data center locations, with 3 more coming online over the next 6 months, and our expanded partnerships bring total accessibility to 253 megawatts, with more to come. We currently expect to exit fiscal 2026 with approximately $400 million of annualized reoccurring revenue. Our $1.44 billion purchase agreement with Dell is essentially fully committed and allocated, and we are in the process of a strategic procurement of an additional $4 billion-$5 billion of compute hardware with multiple OEMs.

Upon completion on the terms we are pursuing, that would be a step change in the scale of this company, anchored by contracted demand and pipeline visibility, and it would allow us to monetize the power commitments we can access today. We are not speculating on hardware. We invest in capacity based on visible trends and continued customer interactions, positioning ourselves ahead of the market rather than reacting to it. We look forward to sharing exciting announcements over the coming months, new customer agreements, expanded capacity partnerships, and further progress on this procurement. Demand has not been our constraint to date. Everything we sold this quarter was contracted before hardware was energized. Inference demand is currently heaviest, and we see no ceiling on it. AI adoption is moving from experimentation into production faster than customers can source capacity.

We compete on the same calls for the same customers as providers many times our size. Quality wins those conversations. We are in active discussions with some of the largest GPU consumers in the world, with requirements ranging from 10,000 to 50,000 GPUs. Those discussions are not yet contracts, and we will announce them only when they are. Here is the math that matters. Given the choice, under the current market conditions, we are preferentially deploy 4 25-megawatt sites over 1 100-megawatt site every time. 4 sites come online faster and in parallel, using a templatized design that has been battle tested across hardware generations and locations. Same capacity, delivered sooner with less risk, repeated throughout the year. That is a multibillion-dollar capacity, and the right question is whether we can run that playbook multiple times a year. We built this company to do exactly that.

A moment on NVIDIA, because the partnership is central to our strategy. Boost Run is an NVIDIA cloud partner and an NVIDIA Exemplar Cloud, one of the smaller number of providers strictly adhering to NVIDIA's reference architecture, and every certification we earn opens the door to the next customer. We work directly with NVIDIA on standardized capacity designs that we deploy consistently across sites, all of which are approved to conform with NVIDIA reference architecture standards. While our platform is architected to support and accelerate at scale, NVIDIA's next-generation technologies sit at the center of our roadmap. Our ongoing collaboration gives us meaningful visibility into what's coming, and together, we are exploring deployments at scale that could meaningfully expand our footprint. Partnerships like this are built on execution, and that remains our focus. Boost Run is built around four core inputs that drive our continued growth. One, the customer. Deliberately diversified portfolio selected on sector, project size, concentration, and creditworthiness.

Reached through direct sales, channel partners, and Boost Run's platform for on-demand access. Second, co-location. We do not own our data centers. We partner with multiple providers and align CapEx with our expanded footprint. Third, access to hardware. A multi-forward-looking approach with our OEM partners and NVIDIA procuring in advance to meet customer demand in a capital-efficient manner. Fourth, finance. Strong relationships with multiple finance partners, allowing us to scale responsibly through prudent use of leverage across short- and long-term structures. We run this company with a focus on operating cash flow by partnering rather than owning data centers. We avoid tying up billions in real estate, eliminate multi-year lead times, and keep capital pointed at revenue-generating hardware. The discipline shows in the numbers.

Adjusted SG&A was roughly $6.4 million against $31.1 million of revenue, about 20% meaningfully below our peer group, with a reconciliation in our supplemental materials. We use the technology we sell. We run AI inference across our own operations, and productivity gains are real. How do we finance profitability growth? Six steps. A rigorous execution plan for every project, with detailed project management and delivery milestones. A required prepayment on every customer agreement, averaging 22% of TCV. Those prepayments are combined with the operating cash flow, and where needed, balance sheet equity to finance each project. That combination is intended to position every project to generate positive project NOI. As contract duration extends, our TCV to CapEx ratio continues to increase to 1.4 and above. We optimize duration to capture the on-demand market, which delivers higher pricing and higher margins through our purpose-built Boost Run platform.

On that last point, currently more than 12% of our revenue comes from short-term on-demand contracts through the Boost Run platform. That is an intentional portfolio construction carrying higher pricing and higher margins, and very few in our space have sustained it at this scale. It is a durable, competitive advantage.

Erik, over to you. Thank you, Andrew, and good morning, everyone.