Los Angeles-based K2 Space is accelerating its path to orbit with fresh venture funding, new defense contracts and a satellite architecture that will be capable of delivering staggering power levels in a single launch.

The company is taking what co-founder and CEO Karan Kunjur described in a recent interview as a “pretty significant contrarian bet against the market.” The premise of the bet goes something like this: The space industry is governed by a single calculus — cost per kilogram of mass. The dollar cost of mass affects how spacecraft are designed, how scientific missions are evaluated and even how entire businesses are planned.

Although the cost per kilogram of mass has declined with the rise of new launch capabilities, like SpaceX’s pioneering work in rocket reusability, spacecraft and mission designers still face egregious mass constraints. As a result, spacecraft have gotten smaller, lighter and more compact. But that doesn’t come without significant trade-offs in power, payload mass and payload volume.

K2 Space is moving in the opposite direction. The company emerged from stealth in March with ambitious plans to design and build massive satellite buses, at never-before-seen costs. Their hypothesis is that next-gen launch vehicles like SpaceX’s Starship will fundamentally change the cost per kilogram paradigm that has ruled with an iron fist for so long — but that to take advantage of this future, we must start planning for it now.

The company is developing two satellite buses: a one-ton payload mass bus called Mega, which can fly on launch vehicles operating today, and a much larger Giga satellite for up to 15 tons of payload, that’s built for super-heavy rockets. These products can “lower the barrier to accessing power, aperture or mass for any application in space,” co-founder and CTO Neel Kunjur explained in a recent interview. (The two co-founders are brothers.)

But it isn’t just the size of the satellites that’s remarkable. Revealed in more detail today, K2 has designed them to operate in a stackable, scalable architecture, so that customers can essentially purchase and operate a high-powered constellation at super-low costs. K2’s products could unlock higher energy orbits, like medium Earth orbit, for companies at price points that are currently prohibitive.

The satellites were designed to maximize launch vehicle fairing volume, too. Stacked together, 10 Mega-class satellites will be able to fit in a Falcon 9 fairing, which would deliver 200 kilowatts of power in a single launch. Even more staggering, 40 Megas will be able to fit into a Starship, which would deliver 800 kW of power in one go.

For reference, ViaSat’s ViaSat-3 telecommunications satellite, one of the most powerful satellites ever deployed in orbit, has 25 kilowatts of power and cost over half a billion dollars to build. In contrast, a 10 Mega-class constellation would cost less than $150 million.

Neel, who spent more than  five years at SpaceX developing avionics for SpaceX’s Dragon spacecraft, said that delivering both high power and high-packing density drove the satellite design.

“Not really many people other than SpaceX are maximizing the capability of Falcon 9,” he explained. “When SpaceX launches a Starlink mission these days, they’re using every kilo of launch capacity they possibly can on Falcon 9. Everybody else is leaving a lot on the table. We wanted to deliver the highest power density per launch possible. At 200 kilowatts deployed per Falcon launch, there’s no other satellite bus manufacturer that comes even close to the amount of power that we can deploy in a single Falcon 9 launch.”

Toward mass abundance

The engineering challenges are substantial. Although there are some major upsides of no longer having to mass optimize every single component — things can be more robust, or made out of heavier (but substantially cheaper) materials — K2 is essentially redesigning a satellite from scratch, even down to the reaction wheel, one of the most fundamental and settled aspects of satellite design (K2’s is one of the largest ever designed). Liberated from entrenched design paradigms that assume mass constraint, around 85% of the spacecraft is vertically integrated, if only because some of the technology doesn’t exist to support the novel satellite architecture.

For that reason, attracting top talent has been key. The company recently brought on Rafael Martinez, who led the design of the original Hall thruster for SpaceX’s Starlink constellation and was director of propulsion engineering at Apollo Fusion, to lead the design of what Karan said will be the highest-powered Hall thruster to be deployed in space by a factor of four. Other notable hires include Ashrith Balakumar, who led the avionics engineering team for SpaceX’s Dragon spacecraft, and Drew Miller, a senior mechanical engineer with experience at SpaceX, Kitty Hawk and Maxar.

K2 has tripled in size since March — growing from a team of six to 18 — and is looking to expand even further to around 40 over the next six months. All this is leading up to the company’s first satellite launch in 2025, for which the launch partner hasn’t yet been announced. Investor interest in K2’s mission has not abated either: The company has also raised an additional $7 million in capital, bringing its total capital raised to $16 million, including an $8.5 million seed round announced earlier this year.

Among the new investors is Alpine Space Ventures, a European fund led by a number of early SpaceX engineers, including Catriona Chambers — who happens to be the person who hired Neel at SpaceX.

The company has also landed three contracts from the U.S. Department of Defense on behalf of different end users, reflecting some interesting traction from defense for K2’s larger platform. The company was awarded the three contracts, which have a total potential contract value of $4.5 million, over the last three months.

“There’s been a real push for resilience in our defense architecture, and historically resilience has come in the form of proliferation, where that proliferation required you going smaller, cheaper, faster,” Karan said. “But unfortunately, there’s a lot of use cases and a lot of end users that actually would like more power than what those small sats offer. The key thing that most of the end users we’ve talked to so far are excited by is being able to have proliferation without sacrificing performance.”

To give a taste of the kind of future the Kunjur brothers are building toward, they described a “dream mission”: launching four or five Mega-class satellites to establish a geostationary communications network around Mars. Scientific missions are deeply constrained in how much power they can send back because they depend on aging Mars orbiters — not to mention any future missions to the red planet.

But that’s just the beginning. A future in space unconstrained by mass is one that has only just begun to be explored.

“Across almost every application, because we’ve been forced to either mass constrain our payloads or volume constrain our payloads or even power constraint them, what we can actually do has been constrained to a large degree,” Karan said. “As a result, the types of missions you can do are more limited.”