Analysis
Stoke Space has closed a $1 billion Series E, TechCrunch reported Tuesday, led by Point72 Ventures and Spark Capital with General Innovation, Glade Brook Capital, US Innovation Technology, Washington Harbour Partners, Woven Capital and Y Combinator also participating. The round pushes the company's total funding to $2.3 billion. Stoke has not flown an orbital mission.
What Stoke is actually building
Stoke Space was founded in 2019 by Andy Lapsa and Tom Feldman, two propulsion engineers who left Blue Origin within a year of each other to build a rocket engineered for reuse from the first sketch, rather than an expendable design retrofitted to survive reentry. The company's Nova rocket carries that idea further than any competitor: both the booster and the second stage are meant to return to the launch site and refly, using an actively cooled, super-chilled liquid-hydrogen shield on the upper stage to survive reentry heating instead of the ablative tiles or heavy metal shielding rivals rely on. Nova Pathfinder, the first vehicle, is rated to carry 3 metric tons to low-Earth orbit and is targeting a first launch in early 2027. A larger variant, Nova Block 2, is aimed at 15 metric tons by 2029. Stoke says it already has launch contracts for multiple Pathfinder flights, though it has not disclosed customers or contract value.
“Nova Pathfinder, the first vehicle, is rated to carry 3 metric tons to low-Earth orbit and is targeting a first launch in early 2027.”
The field it is entering
Reusable launch is no longer a one-company category, and Stoke's pitch only makes sense against the field it is chasing:
- SpaceX -- Falcon 9 reflies its booster routinely but still expends its second stage every flight; Starship is designed for full reuse but has not yet reached a routine reflight cadence. SpaceX itself went public in June at a $1.77 trillion valuation, the largest IPO in history.
- Rocket Lab -- flying Neutron, a partially reusable medium-lift vehicle, as a public company with a multi-year commercial launch backlog.
- Relativity Space -- building the 3D-printed Terran R, still pre-flight after a decade of development and several billion dollars raised.
- Firefly Aerospace -- flying Alpha commercially and developing Eclipse with Northrop Grumman, the most active of the smaller players.
Stoke is the only one of the four betting the whole architecture on both stages coming home. If it works, the operating cost per launch could undercut Falcon 9. If the second-stage reentry shield does not hold up at commercial cadence, Stoke owns the most expensive lesson in the group.
What the number says about capital, not rockets
A $1 billion Series E for a company with zero orbital flights sits inside a pattern Pulse has tracked all year: capital-intensive physical infrastructure is now pricing like software, and other outlets converged on the same read of this specific round:
- Stoke Space — $1B Series E, $2.3B raised to date, zero orbital flights so far (GeekWire)
- Castelion — $800M equity + $250M debt at a $13B valuation, hypersonic-missile startup, August 2026
- Atoms — $1.7B Series A, Travis Kalanick's six-month-old robotics company
- Anduril — raising in similar multiples against defense contracts, not commercial revenue
What connects them is that the checks are being written against a technical thesis and a founding team's track record, the same currency that prices pre-revenue AI labs, just applied to metal instead of models.
What it means for the people writing these checks
For GPs underwriting hardware at this scale, the diligence question is not whether the team can build the engine -- Stoke's founders already have -- it is whether the balance sheet survives the gap between a funded roadmap and a flying, paying customer base. Launch companies have a specific failure mode: the money runs out one test flight before the one that works. Rocket Lab took eight years and dozens of missions before Neutron reached the pad. Relativity is still pre-flight after raising comparable sums. Stoke's investors are betting $2.3 billion says this time is faster.
The counterweight is straightforward. None of this $2.3 billion has yet produced a single dollar of launch revenue, and reusable second-stage reentry -- the hard part of Stoke's bet -- has not been solved at commercial cadence by anyone, including SpaceX, whose Starship upper stage is still years from routine reflight after a decade and roughly $10 billion of development spending that a private company with a fixed cash runway may not be able to match. A slipped Pathfinder date past early 2027 would not be unusual for the industry; it would just be expensive at this valuation.
Nova Pathfinder's first flight is the only test that will actually answer the question this round is pricing.
