Starcloud is raising more money for a future in which AI workloads can run in orbit, not only in data centers on Earth. The startup told TechCrunch it has added a $250 million extension to its March $170 million Series A funding round, bringing its valuation to $2.3 billion.
The new capital is aimed at two connected problems: building bigger space-based compute systems and getting them into orbit. For Starcloud, those goals now depend as much on launch access as on satellite engineering.
Why the new funding matters
Starcloud is developing satellites designed to perform AI inference in orbit. Its next steps include opening a larger manufacturing facility and advancing Starcloud-3, its largest orbital data center spacecraft.
Starcloud-3 is intended to fly on SpaceX's forthcoming Starship rocket. That detail is central to the company's strategy because Starcloud is ultimately built around the possibility that Starship can reduce launch costs enough for orbital inference to compete with terrestrial data centers.
The company is also preparing for scale. Starcloud has already requested permission from the FCC to operate 88,000 spacecraft. That request does not mean all of those satellites are already built or launched, but it shows the size of the network the company wants to be able to pursue.
CEO Philip Johnston is raising capital not only for hardware development but also to secure future rocket transportation. In a sector where the spacecraft cannot create value until it reaches orbit, launch contracts become part of the core business plan.
Launch capacity is becoming the bottleneck
Orbital data center startups face a basic constraint: even if they can design useful compute satellites, they still need reliable and affordable rides to space. The source article notes that launch costs were already one of the biggest challenges for this category, to the point that one startup has decided to build its own rockets.
Starcloud is planning in a market where major launch transitions are underway. SpaceX is planning to phase out Falcon 9, with the Falcon 9 program scheduled to end in 2028, while Starship is larger but still unproven.
Other launch options are also not yet providing the regularity that satellite operators may want. Blue Origin's New Glenn and ULA's Vulcan are not flying regularly, and Rocket Lab's Neutron is not yet on the pad.
That creates planning pressure for companies like Starcloud. The company needs to reserve enough capacity for near-term missions while also positioning itself for a future in which larger spacecraft may depend on Starship availability.
The near-term plan starts with Starcloud-2
Before Starcloud-3, the company is focused on launching two Starcloud-2 spacecraft. These are described as new generation 8 kw compute satellites, and they are slated to fly on rideshare flights in 2027.
Those satellites are expected to perform orbital inference tasks for customers including US government agencies. That gives Starcloud a near-term path to operational work while it continues developing larger systems.
The company is also weighing launch options beyond rideshare. Starcloud is considering buying a dedicated Falcon 9 launch to send up more spacecraft, and it is considering contracts with other launch providers for future missions.
The strategic tension is clear. Rideshare can support early deployments, but a company planning a much larger orbital compute layer may need dedicated and repeatable launch access. That is why the funding extension is tied so closely to launch capacity, not only to satellite production.
Nvidia's role adds technical weight
The funding extension was led by Manhattan West Ventures and included participation from Nvidia and Cisco. Other participants included Benchmark, EQT, Soma, NFX, 776, Cedar Capital, Goanna Capital, and Standard Capital. A person familiar with the deal said Nvidia ponied up $25 million to back Starcloud.
Johnston points to Nvidia's investment as an important signal in the emerging space compute sector. Starcloud is the only company, according to the source, currently operating a Nvidia H100 terrestrial data center GPU in orbit, and it was the first to train a model using it.
That experience matters because most other space GPUs are designed for edge processing. Starcloud is sharing what it learns with Nvidia as the chipmaker develops its first purpose-built GPU for space, the Vera Rubin Space-1 chip.
The chip has not yet been built, but Starcloud hopes to fly it into orbit sometime in late 2028. Johnston said the company's engineers are tracking several design questions, including the relationship between chip running temperature and radiator size, the placement of radiation shielding, and the ruggedizing needed for chips to survive a rocket launch.
A manufacturing base for a larger ambition
Starcloud is still a relatively small company, with 25 employees and growth underway. It is developing production lines at a 100,000 square foot facility in Woodinville, Washington.
The location places Starcloud near where SpaceX and Amazon build satellites for their communications networks. For a company trying to manufacture spacecraft and compete for launch capacity, proximity to a regional satellite production ecosystem may be useful, though the article does not describe specific partnerships tied to the facility.
The larger picture is a race to prove that orbital data centers can become more than a technical experiment. Starcloud has funding, a near-term satellite plan, Nvidia involvement and a long-term dependence on Starship's promise. Its challenge is now execution across several hard fronts at once: chips, spacecraft, manufacturing and launch access.