Max Haot, CEO of Vast, argues that the company can launch the world’s first commercial space station — Haven 1 — in Q1 2027 at a fraction of ISS costs by vertically integrating nearly every component in-house. This 120-minute factory tour in Long Beach is mostly a walkthrough of that hardware: the welded aluminum hull, the in-house hatch, control moment gyros, valves, avionics, micrometeorite shielding, and a full-scale interior mockup with astronaut Drew Feustel. The business case, ISS replacement competition, and long-term artificial-gravity vision come up along the way but take up a smaller share of the conversation than the manufacturing detail.
Haot’s claim is testable within roughly a year of this episode: Haven 1 is slated to launch uncrewed on a Falcon 9, then host a four-person Dragon crew for about two weeks. The reader should come away with Vast’s cost and schedule numbers, its stated advantages over Axiom and Starlab in NASA’s commercial LEO destination competition, and the technical bets — in-house CMGs, consumable CO2 scrubbing, per-module controlled deorbit — that underpin or complicate the pitch.
Vast claims a 10x cost advantage by building the hull in-house
The most concrete claim in the episode is manufacturing economics. Vast mills, rolls, and welds Haven 1’s primary structure from 35 aluminum 2219 panels in about six months for under $5 million, according to Haot. He says competitors buying from the traditional European (Italian) module vendor pay about €55 million and wait four years or more.
| Metric | Vast (in-house) | Traditional vendor (per Haot) |
|---|---|---|
| Primary structure cost | Under $5M | ~€55M |
| Build time | ~6 months | 4+ years |
| Panels per hull | 35 (Haven 1) | — |
| Target build time | 4 months, then under 18 months end-to-end | — |
The hull was pressurized and load-tested at Vast’s Mojave facility, painted, cleaned, and returned to Long Beach for integration. Haot says integration should finish by late summer or early fall 2026, followed by vibration and thermal-vacuum testing at NASA’s Plum Brook station (Glenn Research Center, Ohio), whose vacuum chamber and shake table are large enough for a 15-ton station. Launch campaign at the Cape follows, targeting Q1 2027.
The outsourcing rule was narrow: solar panels from a Spanish supplier and propulsion from Impulse Space, Tom Mueller’s company, which supplies 48 thrusters plus COPV tanks. Everything else — avionics, flight computers, MPPT chargers, batteries, valves, the hatch, the window, the CMGs — is designed, built, and tested in-house, Haot says, “not because we wanted to, because there’s no other choice.” He cites a lack of hatch vendors and year-long lead times on bought components as the driver. The factory runs 25 CNC machines six days a week, 6 a.m. to 2 a.m., and Vast has bought a larger Multiax machine (about 18 months lead time) to cut Haven 2’s panel count to roughly 15 and reduce welding.
Haven 1 is sized to the Falcon 9 fairing and pitched as the third-largest object in space
Haven 1 is a single-module station with one docking port, 55 cubic meters of pressurized volume (about five times the free-flyer Dragon missions), 10 mid-deck-locker-equivalent payload slots, and a 1.2-meter curved window that Haot calls the biggest single pane in space. It launches uncrewed on a reusable Falcon 9 using the full fairing, at about 15 tons. Haot estimates that makes it the third-largest object in space by mass, after the ISS and China’s Tiangong.
The nominal mission is two days transit, 10 days on station, two days back, with crews allowed up to 30 days. Haot frames the duration as the key upgrade over Inspiration4-style free flyers: about half of people get space sickness and need two to three days to recover, leaving little productive time on a three-to-five-day trip. Dragon serves as both transport and lifeboat; Vast must convince SpaceX with data that the station is safe before a crew is committed. Vast resells the transport as part of its mission package, including SpaceX suit fitting and training.
Connectivity is a differentiator Haot stresses: Haven 1 will carry Starlink laser terminals connecting directly to passing satellites, giving gigabit, low-latency links for crew and payloads — he says better than the ISS’s GEO-relayed link, whose bandwidth is decent but whose latency is the real constraint. Vast is the first space station provider to sign a Starlink deal with SpaceX, per Haot.
The station is designed to deorbit itself, module by module
Vast flew a 500 kg demo satellite on November 2, 2025, and Haot says it achieved 100% mission success with no anomalies, testing everything except the pressurized volume, hatch, docking, and window. The finale was a controlled deorbit: a 16-minute burn of an Impulse 5N thruster after lowering perigee, targeting a few-hundred-kilometer box near Point Nemo, which it hit slightly off-center.
Haot ties this to a policy argument. NASA plans to deorbit the whole ISS in 2030 using a dedicated SpaceX-built vehicle, a budget line he calls irresponsible. Vast’s approach is that every module carries its own propellant and propulsion to deorbit itself at end of life. He invokes Skylab, which failed to deorbit controllably and scattered debris over Australia, as the precedent to avoid, and notes the FCC requires proof that ground casualty risk falls within a threshold before launch.
Haot gives three reasons the ISS ends in 2030: the Russia partnership is geopolitically undesirable, modules are 20–30 years old with leaks on the Russian side, and cost — about $4.5 billion a year of NASA’s $25 billion budget, which the agency hopes to cut below $2 billion.
NASA’s LEO competition is the near-term market; in-space manufacturing is the option value
Haot is blunt that the current market is sovereign governments. Vast is competing in NASA’s Commercial LEO Destinations program against Axiom Space and Starlab, with Blue Origin a possible entrant; SpaceX is a mission partner, not a module competitor in his framing. He says winning NASA brings the other ISS partners — Japan, Europe, Canada — who have said they will buy services from whoever NASA selects. He also cites growing demand from Saudi Arabia, the UAE, Poland, and Turkey, and estimates self-funded private missions at 10–15% of the business.
His profitability argument: even if NASA’s spend halves, a low-cost approach can be profitable on today’s government market alone. That robustness is the point — it lets Vast survive however long it takes for a killer application in space manufacturing to emerge. He is most excited about drugs that microgravity makes possible, plus stem-cell organoids and cancer-treatment patches, and argues the ISS’s throughput and government queueing rules throttle that research. He also names foreign governments buying their own modules, Space Force and allied military habitats, and lunar or Mars habitats as adjacent businesses.
“The moment we get one that works then we are weaning off the government and we are moving into the manufacturing era where these modules are no longer research lab they are you know factories basically.”
Roadmap runs from one module to a seven-Starship spinning station
Haven 2 is the multi-module station intended to match ISS power and life support, built as derivatives of Haven 1 at a target of at least three modules per year. The long-term concept, which Vast calls the Stick, is a 7-meter-diameter, 105-meter-long spinning station assembled from seven Starship launches, rotating at four rpm to give 1g at each end, with lunar and Mars gravity bands and microgravity science in the center. Haot’s rationale: 1g is proven for a lifetime on Earth, zero-g is tolerable for perhaps a year, and nobody knows whether lunar or Mars gravity sustains health — an artificial-gravity station is how you find out.
In-house subsystems carry the safety case and the schedule
The tour’s technical core is the argument that vertical integration is also a safety and iteration strategy. Vast developed its own control moment gyros — Haot says no competitor has — because no commercial CMG is sized for a station; six fly on Haven 1, and they have been spinning on test stands for nearly three years to qualify bearing life. Using CMGs instead of thrusters keeps attitude control on solar power rather than finite propellant.
Valves are the recurring failure point Haot calls out, referencing Starliner’s valve problems. Vast buys where it can (a Marotta valve) and designs the rest in-house, including oxygen and pressure-relief valves. The redundancy standard: two-fault tolerance to loss of crew, one-fault tolerance to loss of mission. Propulsion uses 48 thrusters with duplication so a single valve failure is survivable.
Micrometeorite protection uses a multi-layer Whipple shield over the pressure hull. Vast fires roughly 1 mm aluminum projectiles at 7.5 km/s — Haot’s comparison is about 20 times a handgun bullet — at a test article showing the outer layer spraying into a cone that the ballistic layer stops before the pressure wall. Vast has bought its own orbital-velocity gun because outside test capacity was insufficient. Every electronics box goes through thermal cycling (Haot cites swings of +100°C to −100°C every 90 minutes) and vibration tables in-house.
Astronaut input reshaped the interior, and one design was overruled
Drew Feustel, Vast’s lead astronaut with 226 days in space across two shuttle missions and a six-and-a-half-month ISS stay, walks through the mockups and describes where astronaut experience changed the design. The clearest example: engineers originally built a cover over the inter-module ventilation duct for aesthetics. Feustel rejected it because the crew must be able to throw the duct aside and close the hatch fast in a fire or depressurization. It was redesigned as a flexible duct.
Other specifics: four sleep quarters (two overhead, two in the floor) with an inflatable bladder system Feustel helped design that presses the sleeper against a soft wall to mimic a weighted blanket, replacing the bungee-cord tethering astronauts use on the ISS. Lighting shifts from blue-rich morning light to red night light to protect circadian rhythm from the 35-minute orbital day-night cycle. The toilet flies on a pallet and is moved into Dragon for use, with Dragon’s own toilet reserved for ascent and descent. Feustel stresses that Haven 1’s interior is roughly ISS-module diameter (about 4 m) and length (10–11 m), but with hardware and stowage tucked into the walls rather than “growing from the inside” as on the 25-year-old ISS, and with a planned three-year vehicle refresh cycle.
“You can’t ever trade aesthetics for safety, right? Safety is paramount. That’s that’s the number one thing. And the hierarchy is you save the crew, you save the spacecraft, you save the mission.”
Competition is thin, and the binding constraint is time
Haot names time as Vast’s main competitor. No company has yet operated a commercial space station, so he counts zero current entrants and expects Vast to be “one of one” if Haven 1 flies first. He puts more than $1 billion invested to date, with more spent before Haven 1 flies, as the barrier that makes the position hard to follow. On China, he says the commercial habitat scene is still government-dominated, unlike launch, and that docking standards diverge — NASA published the International Docking Standard openly, Russia uses its own, and China’s may draw on the NASA standard.
The unresolved tension is that Vast’s cost and schedule claims are self-reported and pre-launch. The hull cost, the six-month build, the CMG qualification, and the Q1 2027 date all get their first real test at Plum Brook and the Cape. If Haven 1 launches and a Dragon crew docks on schedule, the NASA competition and the module-sales pipeline to foreign governments become live; if the schedule slips, the “profitable on today’s government market” argument is what keeps the company funded while it waits for a manufacturing breakthrough that Haot himself says could be 2, 5, 10, or 15 years away.