K2 Space is about to do something that sounds like it shouldn’t be that big of a deal but actually might reshape how we think about satellites. The company, founded by brothers Karan and Neel Kunjur who both cut their teeth at SpaceX, is launching a spacecraft called Gravitas that can generate 20 kilowatts of continuous power. That doesn’t sound like much until you realize most satellites barely scratch a few kilowatts.
The timing is interesting because we’re at this weird inflection point where space is transitioning from being a novelty to being actual infrastructure. Starlink isn’t some sci-fi dream anymore. Amazon’s building a competing constellation. The Pentagon just committed $185 billion to a satellite-based missile defense system. Suddenly, people care about what’s happening up there.
The Power Problem Nobody Talks About
Here’s the thing about satellites that most people don’t understand. You can’t just slap more computers on a spacecraft and call it a day. Everything needs juice. Sensors need power. Communication equipment needs power. Data processing needs power. Propulsion definitely needs power.
Most existing satellites are basically running on pocket change when it comes to electrical capacity. ViaSat-3 can squeeze out more than 25 kilowatts, and Starlink V2 hits around 28 kilowatts, but these are the exceptions. They’re also incredibly expensive. K2’s bet is that there’s a sweet spot where you can offer genuinely useful power levels without requiring traditional contractors’ pricing models.
Gravitas carries a 40-meter solar panel wingspan when fully deployed. That’s huge. That’s the kind of solar collection area you need to actually power something meaningful at 20 kilowatts continuously. The company is also flying what they claim will be the most powerful electric thruster ever tested in orbit. That’s not just flexing. That thruster needs serious electricity to do its job, and if it works as planned, K2 can demonstrate they’re serious about this whole “high power in space” thing.
Why This Launch Matters More Than You’d Think
K2 raised $425 million and hit a $3 billion valuation in December 2025. That’s a lot of confidence from investors who know the space industry well. But Gravitas is their first actual mission. Everything before this was engineering and promises. There’s a real possibility something goes wrong on launch, during deployment, or when the payload modules fire up.
Karan Kunjur is being realistic about this. He’s framed success in tiers. First tier is just getting the satellite to orbit and generating power. Doesn’t sound easy when you’ve built 85% of the components in-house. Second tier is running the various payload modules from Department of Defense customers and others, plus testing that massive electric thruster. Third tier would be using that thruster to raise Gravitas thousands of kilometers into a higher orbit. Each tier is basically “okay, that worked, can we do something harder?”
The company plans to launch eleven more satellites over the next two years. By 2028, Kunjur expects K2 to be the satellite supplier for commercial networks that actually need real power. That’s an aggressive timeline, but not insane if Gravitas performs well.
The Starship Problem
There’s an elephant in the room here, and its name is Starship. K2’s original pitch was basically “SpaceX will build an enormous rocket that makes launch costs trivial, and we’ll be positioned to take advantage.” That rocket is still being developed. It’s not flying commercial missions yet. It might be years away.
So K2 has to exist in a world where Falcon 9 costs roughly $7.2 million for a launch slot. That’s the current reality. Their satellites cost around $15 million. Traditional contractors charge way more for less capable hardware. It’s not a perfect position, but it’s defensible. Once Starship actually starts flying regularly and those launch costs drop to something like $600,000, suddenly you can fit way more massive satellites on each mission.
Kunjur told TechCrunch they’re already designing 100 kilowatt satellites. The designs are apparently complete enough that they have mockups taking up an entire factory floor. That’s the bet. Build the infrastructure and expertise now so that when the rocket industry catches up, you’re already in position.
What This Actually Means
Communications networks are about to get a lot more interesting. More power means more throughput. It means harder to jam. It means ground stations can be cheaper because they don’t need to be as sensitive if the satellite is screaming data at higher power. Amazon LEO, Starlink, and whatever else gets built will eventually want this kind of capability.
Data processing in orbit stops being theoretical once you have satellites that can actually power serious computing equipment. The hyperscalers see this coming. Amazon, Google, Meta. These are companies that understand data gravity and latency. Having processing happen at the edge of space instead of on the ground starts mattering when you’re managing continental-scale operations.
The military implications are obvious too. Better communications. Better sensors. Better redundancy. The $185 billion defense constellation isn’t happening without something like Gravitas-class technology becoming commonplace.
K2 is betting that power is the constraint everyone’s been underestimating, and that once you remove that constraint, the market reshapes itself. They might be right. They might launch Gravitas and run into a problem nobody expected. Either way, we’ll find out in the next few weeks what happens when you take the space industry’s power problem seriously enough to build a company around it.
