Building the First Data Centers in Space
Philip Johnston, CEO of StarCloud, discusses how the company is building data centers in space to solve Earth's energy bottleneck for AI by leveraging unlimited solar power and declining launch costs. The company launched its first satellite in November 2024 with high-power NVIDIA GPUs and plans to deploy thousands of satellites providing 20 gigawatts of compute capacity.
Summary
Philip Johnston, co-founder and CEO of StarCloud, explains the company's mission to build data centers in space. Founded on January 1, 2024, StarCloud became the fastest-growing unicorn in YC history, raising $170 million led by Benchmark just 17 months after Demo Day.
The core idea emerged from Johnston's observation at SpaceX's Starbase in early 2023, where he realized that reusable Starship rockets could eventually reduce launch costs by 10-100x and increase capacity by 1,000x. He recognized that space-based solar and data centers would make economic sense in this future. Initially, the company explored space-based solar but pivoted when calculations showed that data centers had a more realistic break-even launch cost of $500 per kilogram (versus $50 per kilogram for solar).
StarCloud One, launched in November 2024, marked a major milestone. The satellite contains five GPUs including an NVIDIA H100—something many believed was physically impossible due to thermal and radiation challenges. The team used immersion cooling with phase-change material to manage heat in the vacuum environment. The satellite deployed successfully, achieved first contact in 12 hours, underwent two weeks of commissioning, and successfully ran inference, training, and model fine-tuning operations in orbit.
The company's roadmap includes StarCloud Two (10 kilowatt spacecraft for government/military customers), StarCloud Three (200 kilowatt spacecraft for commercial data centers), and a proposed constellation of 88,000 satellites in sun-synchronous orbit providing 20 gigawatts of compute capacity—20 times the entire US power grid and far exceeding the largest terrestrial data centers at ~1 gigawatt each.
Major engineering challenges include thermal management and radiation tolerance. StarCloud developed radiators that are 10x less massive and 500x cheaper than ISS radiators per watt of heat dissipation. For radiation, the team extensively tests components at particle accelerators to understand failure modes and implements software and hardware shielding solutions. They partnered with NVIDIA on a specialized Ruben space chip designed for the space environment.
Funding was extremely challenging initially—the team was rejected by approximately 100 VCs before raising their first $2 million seed round. The Series A was also difficult, with 20+ rejections before closing with Benchmark. However, sentiment shifted dramatically as SaaS valuations declined and it became clear that regulatory restrictions on terrestrial data centers (particularly in New York and other states) were accelerating.
Johnston's background combines software engineering, physics, mathematics, and McKinsey consulting experience on satellite missions—not traditional space engineering. He deliberately recruited world-class space engineers (including co-founders from SpaceX and NASA) before determining the exact business idea, following YC advice to solve for technical talent first. The current 20-person team remains extremely selective in hiring.
The company addresses critical national security needs, as AI compute capacity has become essential for Western competitiveness. StarCloud positions itself as a solution to regulatory headwinds against terrestrial data centers while providing unlimited clean solar energy in space.
About this episode
<p>Philip Johnston is the co-founder and CEO of Starcloud, the company building data centers in space. </p><p><br /></p><p>In November 2025, Starcloud launched an Nvidia H100 GPU into orbit and trained the first large language model in space. They've since raised $200 million, hit a billion-dollar valuation just 17 months after YC demo day, and filed with the FCC to deploy 88,000 more satellites. </p><p><br /></p><p>In this episode, Philip walks us through their wild origin story, the engineering challenges behind the Starcloud-1, why they booked a SpaceX launch before they even knew what they were building, and how data centers in space make sense both economically and politically.</p>
Key Insights
- StarCloud pivoted from space-based solar to data centers because the break-even launch cost was $500/kg for data centers versus $50/kg for solar, making the former far more realistic economically.
- The team successfully ran NVIDIA H100 GPUs in orbit using immersion cooling in phase-change material—a solution that nobody had previously considered viable for space applications.
- Johnston deliberately recruited world-class space engineers (from SpaceX and NASA) as co-founders before crystallizing the specific business idea, relying only on the insight that launch costs would decline dramatically.
- Investor sentiment toward space and hard tech shifted rapidly starting in late 2023, driven by declining SaaS valuations and regulatory bans on data centers in major US cities like New York.
- StarCloud's custom radiators achieve 10x better mass efficiency and 500x better cost efficiency than ISS radiators, proving that existing space technology can be engineered for commercial viability.
- The company spent only $2 million deploying StarCloud One (including launch costs) compared to traditional aerospace primes' estimate of $75-100 million, demonstrating the cost advantages of startup approaches to space engineering.
- Regulatory restrictions on terrestrial data centers—often based on water and power consumption concerns that Johnston argues are not scientifically grounded—are creating urgent political demand for alternative compute infrastructure.
- StarCloud's proposed 88,000-satellite constellation could provide 20 gigawatts of compute capacity, equivalent to 20 times the entire US power grid, fundamentally shifting the scale of available computing infrastructure.
Topics
Transcript
First thing every space company should do is book the first available launch they can. Before they built the thing that they're going to launch. Before they built the thing, before they probably even know what they're going to launch. Booking a launch is such a good forcing function for a space company. So we founded the company January 1st, 2024. January 2nd, we booked the first available SpaceX rideshare launch. And we were like, okay, something is going to be on that rocket. I'm not sure 100% sure what it's going to be at this point, but something is going to be on there. Sounds like the same advice for all the software companies. Just got to keep launching…
Full transcript available for MurmurCast members
Sign Up to AccessMore from Y Combinator Startup Podcast
Waymo Co-CEO Dmitri Dolgov: "Move Fast And Ship Safely"
Waymo Co-CEO Dmitri Dolgov presents seven technical lessons learned from building and deploying autonomous vehicles at scale, emphasizing the critical difference between demos and production systems, the importance of architectural choices aligned with safety requirements, and the necessity of a comprehensive AI ecosystem including simulators and rigorous evaluation frameworks.
Patrick Collison: "What If You Succeed?"
Patrick Collison, CEO of Stripe, discusses his journey from dropping out of MIT to founding Stripe, emphasizing the importance of grounding decisions in real customer problems and the value of deep knowledge even in an AI era. He shares data showing unprecedented growth in new business formation and argues that this is the best time ever to start a company, with AI creating more decentralization rather than concentration.
Jeff Dean: The 1% Rule for Building in AI
Jeff Dean discusses AI progress, hardware specialization, and the future of agent-based systems. He argues that inference optimization and context engineering are key frontiers, while emphasizing that taste and problem selection—not just technical execution—will differentiate successful AI-native founders from competitors.
Alexandr Wang: “This is a Once-in-a-Civilization Opportunity”
Alexandr Wang discusses his journey from math olympiad competitor to founding Scale AI, emphasizing the importance of developing independent conviction about the future while working on AI as a once-in-a-civilization opportunity. He shares insights on building Meta's AI research lab, the shift from human intelligence as a bottleneck to vision and ambition as the scarce resource, and practical approaches to agentic systems.
Blake Scholl: Breaking the Supersonic Ban
Blake Scholl, founder of Boom Supersonic, discusses building the first independently developed supersonic jet (XB-1), breaking regulatory barriers to legalize supersonic flight in the US, and how applying software development principles to hardware manufacturing can reduce costs and accelerate iteration in aerospace. He emphasizes the importance of working on problems you love, building in public to inspire others, and challenging the assumption that good ideas always get built quickly.