Starcloud, which launched as Lumen Orbit, builds data centers in space, first to give other satellites GPU compute, then to absorb the energy demand AI is creating on the ground. Philip is a second-time founder who worked on satellite projects for national space agencies at McKinsey. In November 2025, 21 months after founding, Starcloud launched Starcloud-1 carrying an Nvidia H100, a GPU 100 times more powerful than any previously operated in space. It became the first company to train an LLM in orbit. In March 2026 it raised a $170M Series A led by Benchmark, the fastest unicorn in YC history at 17 months from demo day. Its second satellite launches in October 2026.
Philip Johnston is the co-founder and CEO of Starcloud, which builds data centers in space. He is a second-time founder and previously worked at McKinsey on satellite projects for national space agencies.
A Y Combinator S24 company putting data centers in orbit, starting with GPU compute for other satellites and moving towards the energy demand created by AI. Its case rests on abundant solar energy, radiative cooling and no permitting queue.
Launched in November 2025, 21 months after the company was founded, carrying an Nvidia H100, a GPU 100 times more powerful than anything operated in space before. Starcloud became the first company to train an LLM in orbit and the first to run a version of Gemini there.
A $170M Series A led by Benchmark in March 2026, which made it the fastest unicorn in Y Combinator's history, 17 months after demo day. Earlier investors include NFX, Nvidia and In-Q-Tel.
A second satellite launching in October 2026 with 100 times the power generation of the first, and on-orbit inference on satellite radar data.
Energy and cooling. Space offers abundant solar energy and radiative cooling, with room to grow without the land, grid and permitting limits that hold back terrestrial builds.
