Starcloud now sits at a $2.3 billion valuation after adding $250 million to its Series A round. That is the strangest part. A company that wants to put data centers in orbit just convinced investors it is worth more than most airlines.

The money came fast. Starcloud raised $170 million back in March. Now, less than six months later, it has another $250 million on top. Manhattan West Ventures led the extension. Nvidia and Cisco joined in. A person familiar with the deal told TechCrunch that Nvidia put in $25 million on its own.

You can feel the excitement in the room when you read the numbers. This is not a small bet. It is a bet that the future of computing lives above our heads, not in warehouses on the ground.

The factory that opens next week

CEO Philip Johnston took to LinkedIn to share the news. He did not hide the momentum. “Starcloud is set to open a 100,000-square-foot satellite manufacturing center next week.”

That factory sits in Woodinville, Washington. It is near where SpaceX and Amazon build their own satellites. Starcloud plans to use the space to ramp production to 100 satellites per week.

Think about that for a second. One hundred satellites a week. The company currently has about 25 employees and is growing. It is building production lines for its next spacecraft, called Starcloud-3, which is designed to fly on SpaceX’s Starship rocket.

The math adds up to something wild. Starcloud has now raised about $450 million since it was founded in 2024. That is a lot of money for a company that has not yet launched a full data center into orbit.

The rocket problem nobody is ignoring

Here is where the story takes a turn. Starcloud needs rockets to get its satellites up there. And rockets are getting harder to book.

SpaceX is transitioning from its workhorse Falcon 9 to the bigger Starship. That shift is squeezing launch capacity. Falcon 9 rideshare slots are not available beyond late 2028 or early 2029.

Johnston said it plainly. Launch availability is the company’s immediate constraint. “Obviously if we can’t book any SpaceX launch capacity in 2029, that will be challenging for us,” he said.

So the company is using part of this new money to secure launch allocation. It is also working with Nvidia on hardware that can survive the radiation and extreme conditions of space. Nvidia unveiled a computing system in March called the Space-1 Vera Rubin Module for exactly this kind of work.

The long-term plan is massive. Starcloud wants to deploy 88,000 satellites with a combined 20 gigawatts of orbital compute capacity. That is more power than many countries use. All of it floating above us.

Why this matters now

You might wonder why investors are writing checks this big for something that sounds like science fiction. The answer is simple. AI needs compute. And compute needs power. And power needs space.

Data centers on the ground are hitting limits. They use huge amounts of electricity. They generate heat. They need cooling. Putting compute in orbit sidesteps some of those problems. Space is cold. Solar power is abundant. There is no real estate bill.

Nvidia’s involvement signals something important. The chip giant sells hardware. It is backing the infrastructure that will run its chips in new places. Cisco’s participation points the same way. These are not small players testing the waters.

But there is risk. Launching and maintaining data centers in space is not cheap. It is not easy. The technology is new. The logistics are untested at this scale. And the rocket bottleneck is real.

Starcloud’s Starcloud-3 spacecraft is designed to carry 200 kilowatts of compute capacity. That is a big jump from its earlier satellites, which combined for about 16 kilowatts. The company is betting that bigger spacecraft on bigger rockets will solve the math.

The quiet part

There is a moment in this story where you stop and think. Not about the money. Not about the rockets. But about what it means to put so much computing power above our heads.

Starcloud is not alone in this race. Other companies are looking at orbital compute. But few have moved this fast or raised this much. The company doubled its valuation in less than six months. That is a signal. Investors see something here.

The factory opens next week. The satellites will roll off the line. The rockets will need to be booked. And somewhere between the factory floor and the launch pad, the whole plan either works or it does not.

Johnston and his team are betting it works. Nvidia and Cisco are betting it works. Manhattan West and the other investors are betting it works.

The world will find out soon enough.