The YC S24 startup began with two engineers, some robotic arms, and a clothes dryer for a desk. Now it is building refueling depots so spacecraft can top up in space instead of carrying every drop from launch.
A satellite can be perfectly healthy and still be dead. Its electronics work, its solar panels catch the sun, its instruments are calibrated - but the fuel tank is empty, so it drifts off station and joins the slow parade of junk in orbit. Most spacecraft die this way: not broken, just out of gas. Spaceium, a company that went through Y Combinator's Summer 2024 batch, is trying to make that a solvable problem instead of a terminal one.
The pitch is almost suspiciously simple. Spacecraft carry all the fuel they will ever use from the moment they leave the launch pad. Every kilogram of that propellant is a kilogram of camera, antenna, or science instrument they cannot carry instead. Spaceium's idea is to break that trade-off by putting the fuel where the mission is - in orbit - and letting spacecraft refuel the way a car pulls into a station on a long drive. The founders like to call the eventual network a "space superhighway," with service stations along the route.
Illustrative split of launch mass. Fuel a spacecraft must carry for its whole mission is mass it can't spend on payload.
Less fuel launched means more payload, longer life, or farther travel. Directional, not to scale.
Spaceium's work rests on two pieces of hardware and one plan to combine them. The first is a cryogenic fuel storage system that, according to the company, can hold multiple metric tons of propellant in orbit for up to ten years without boil-off. That last part matters more than it sounds. Cryogenic fuels are cold and eager to evaporate; keeping them liquid and contained over years, in vacuum, is the hard engineering problem that separates a fuel depot from an expensive leak. CEO Ashi Dissanayake has described it plainly.
"We developed a very unique system where we can store the fuel for longer periods of time, which was actually not done before." Ashi Dissanayake, Co-Founder & CEO
The second piece is a robotic arm - and the actuator that drives it. In late 2025, Spaceium flew that actuator to orbit aboard SpaceX's Transporter-15 rideshare mission and reported a verified rotation accuracy of 0.003 degrees, which it describes as roughly 70 times more accurate than robotic arms currently operating in space. Dissanayake put the number in human terms: paired with a five-meter arm, that precision translates to less than a millimeter of movement at the far end. For a machine trying to mate a fuel line to a docking spacecraft, sub-millimeter is the difference between a fill-up and a fender-bender.
The third piece is the station itself: an unmanned depot that combines storage and arm to refuel docking spacecraft, and, over time, to inspect and repair them. The plan is modular - the same building blocks adapted for different orbits and jobs. As Dissanayake summarized it, "We have our robotic arm that we can use to transfer cryogenic fuel, non-cryogenic fuel, as well as to provide repairs."
Spaceium did not start in a clean room. In 2023, co-founders Ashi Dissanayake and Reza Fetanat were working out of a small apartment in Ottawa, prototyping with disembodied robotic arms and, by their own account, using a clothes dryer as a desk. The two had met at the University of Ottawa, where they bonded over space hardware and built a rocket that reached about 30,000 feet, along with propulsion systems and recovery parachutes.
That background - actual rockets, actual propulsion, actual payload integration - is part of why investors took the leap on infrastructure that does not exist yet. The company is structured as a U.S. entity with a Canadian subsidiary, a nod to those Ottawa roots, and now operates out of Long Beach, California - a stretch of coastline that has quietly become one of the densest concentrations of small aerospace companies in the country. Dissanayake is CEO; Fetanat is CTO. The team remains small, in the range of six people.
Six people is a useful number to sit with. It is not a research lab with a hundred engineers and a decade of runway. It is a group small enough to fit around one table, betting that focus beats headcount on a problem that larger, better-funded organizations have circled for years. Spaceium's stated values read like a reaction to that constraint rather than a poster on a wall: a bold mission, relentless innovation, and fast execution. In practice that has meant shipping a real piece of hardware to orbit early - one actuator on a rideshare - rather than waiting to unveil a finished station. The demo was small on purpose. A single component in the real environment is falsifiable in a way that a rendering never is.
In January 2025, Spaceium announced a $6.3 million seed round, oversubscribed, led by Initialized Capital, with a long list of funds and angels alongside it - Zeno Ventures, Olive Tree Capital, DTX Ventures, Crosscourt Ventures, Alumni Ventures, Spacedust Ventures and others among them. Add the earlier Y Combinator investment and the company has raised roughly $6.4 million to date. The stated use of funds is unglamorous and exactly right for the stage: demonstrate the technology in orbit, and grow the team.
Disclosed rounds, in USD.
Here is the part that makes hardware skeptics raise an eyebrow: Spaceium says it already has customers. Before a single station has flown, the company reports bookings for refueling missions in 2029 and 2030, and a pipeline beyond that. Its buyers are the working population of orbit - satellite operators who want to extend a spacecraft's life, orbital transfer vehicle companies moving payloads between orbits, and government and defense programs that would rather not abandon a working asset because it ran dry.
The near-term proof point is a partnership with Space Machines Company on what the two describe as a first-ever cryogenic refueling demonstration in space, with Spaceium's storage and transfer technology hosted on SMC's platform. The roadmap after that is specific: an automated refueling station targeted for 2027 carrying multiple metric tons of fuel, scaling toward tens of tons on station by the end of the decade.
Booking revenue years before delivery is a familiar move in hardware, and it cuts both ways. It signals that operators believe the problem is real and worth reserving capacity against. It also means the company is now on the hook to build the thing it sold. The business model underneath is straightforward B2B infrastructure: sell refueling by the mission, add life-extension and repair as recurring services, and price competitively against the alternative, which for a satellite operator is often no alternative at all - the spacecraft simply ages out. Spaceium says its planned pricing is "really competitive" against industry options, though depot economics in orbit remain, for everyone in the field, largely untested.
"We have our robotic arm that we can use to transfer cryogenic fuel, non-cryogenic fuel, as well as to provide repairs." Ashi Dissanayake, Co-Founder & CEO
Spaceium is not alone in noticing that orbit needs a gas station. Orbit Fab has been marketing refueling interfaces and "gas stations in space" for years, and Astroscale has built a business around on-orbit servicing, including a U.S. Space Force contract and planned GEO refueling operations. What Spaceium is betting on is the combination: long-duration cryogenic storage, a notably precise arm, and repair as a service, all in one modular station. Whether that bundle wins is unproven - but the market it aims at is the connective tissue of the entire space economy.
The flashy space stories are about rockets and Mars. Spaceium is building the part in between: the logistics, the top-ups, the repairs that let expensive machines keep working instead of becoming debris. It is early, and much of the plan is still a plan. But the thesis is durable - as long as spacecraft burn fuel, someone will want to sell it to them where they are. If that turns out to be a real market, it will have been built by people who were once designing around a clothes dryer, and who decided the most useful thing they could invent was a place to stop and fill up.