Summary
Y Combinator’s Light Cone podcast sits down with Philip Johnston, co-founder and CEO of StarCloud, to explore why the company is building data centers in space to solve the energy constraints AI is creating on Earth. Johnston explains that terrestrial power capacity is increasingly the binding constraint on new data center construction — with bans already enacted in New York and expected in several other states — while solar energy available in low Earth orbit is effectively unlimited. The economics hinge on rapidly declining launch costs driven by SpaceX’s Falcon 9 and the forthcoming Starship program, which Johnston argues could increase launch capacity by orders of magnitude within a few years.
StarCloud raised $170 million in a round led by Benchmark and became the fastest-growing unicorn in YC history just 17 months after Demo Day — a remarkable turnaround from being rejected by more than 100 VCs during the company’s early fundraising. Johnston recounts how the idea evolved from space-based solar power (abandoned after discovering 95% transmission losses) to co-locating compute directly in orbit, where the power is generated.
The conversation covers the engineering rationale for why data centers are the ideal first space workload — no re-entry required, unlike asteroid mining or manufacturing — the founding team’s roots at SpaceX and NASA, and the practical startup advice of booking a rocket launch before you’ve decided what to put on it as a forcing function for execution discipline.
📺 Source: Y Combinator · Published August 05, 2026
🏷️ Format: Interview
