The capsule landed with a thud on the red Australian outback, kicking up dust at a test range most people have never heard of. Inside, carefully cushioned against the violence of reentry, sat vials of crystallized ritonavir—an HIV drug that had spent weeks floating in the absolute stillness of low Earth orbit.
It was February 27, 2025, and Varda Space Industries had just pulled off something that sounded almost quaint in its ambition: making pharmaceuticals in space and bringing them home. Not for science fiction reasons. For chemistry reasons. Because sometimes gravity gets in the way.
That landing at South Australia's Koonibba Test Range marked more than a technical milestone, though it was the first commercial spacecraft to touch down at a commercial spaceport. It validated a premise that materials scientists have kicked around for decades, one that always felt tantalizingly close but maddeningly impractical: certain things can only be manufactured properly without gravity's interference.
Now the race to industrialize that premise is accelerating, and it's being led by a cohort of startups you've probably never heard of. They're building reentry capsules, orbital furnaces, and—in one case—autonomous space stations with no humans allowed. The target markets are niche but lucrative: drugs that crystallize into more effective forms in microgravity, semiconductor substrates grown with near-perfect atomic precision, optical fibers free of the microscopic flaws that plague Earth-bound production.
The global space economy hit $613 billion in 2024, according to the Space Foundation's report published in July 2025. In-space manufacturing represents a rounding error in that total today. But McKinsey projected in April 2024 that the broader space economy could reach $1.8 trillion by 2035, and the infrastructure being built right now is a bet that a meaningful slice of that growth will come from making things up there and bringing them back down here.
What's changed? Launch costs have cratered. Reentry licensing, once a byzantine nightmare, is maturing into something approaching a repeatable process. And the technical proof points are stacking up—slowly, imperfectly, but undeniably. The question has shifted. It's no longer whether microgravity manufacturing works. It's who can build the return logistics fast enough, and whether anyone on Earth will actually pay for what comes back.
Three Approaches, One Infrastructure Layer
The companies staking claims in this nascent industry have coalesced around three rough architectures, each with its own trade-offs.
Varda has the clearest operational track record. Since its first FAA-licensed commercial reentry in February 2024, the company has executed four missions through January 2026. The W-series capsules—W-2, W-3, W-5—have all recovered payloads at Koonibba under an agreement that runs through 2028 and covers up to 20 additional reentries. Each capsule uses C-PICA heat shields, an ablative material originally developed by NASA that Varda now manufactures in-house. The shields have proven their worth, protecting delicate biological payloads through deceleration peaks around seven times the force of gravity. In June 2025, the FAA granted Varda an expanded reentry vehicle operator license—the first of its kind—which streamlines future approvals considerably.
Across the Atlantic, UK-based Space Forge is pursuing something different: reusable platforms that stay in orbit longer and return only when they're carrying something worth bringing home. The company's ForgeStar-1 satellite, launched in June 2025 on a SpaceX rideshare, houses a plasma furnace that hit temperatures exceeding 1,000°C in orbit last December. CEO Joshua Western called it proof of "the essential environment for advanced crystal growth" on a dedicated commercial platform. Space Forge is developing Pridwen, a deployable and reusable heat shield that completed zero-gravity deployment tests in 2025, alongside a soft-capture return system called Fielder. A £22.6 million Series A in May 2025 and a July partnership with Intuitive Machines to integrate U.S. return capability signal the company's intent to bring semiconductor materials back from orbit in future missions.
Then there are the smaller players testing adjacent approaches. Inversion Space received an FAA reentry demonstration license in October 2024 for its Ray capsule, focused on defense logistics and rapid return. Outpost has completed flight tests of an autonomous paraglider reentry system and secured a STRATFI award from the Defense Department in August 2024, though it hasn't yet reported an orbital return. SpaceWorks offers a family of RED reentry capsules with parafoil precision landing for payloads ranging from 4 kg to 50 kg. France's Atmos Space Cargo is developing an inflatable heat shield concept. Spain's Orbital Paradigm has ESA backing but remains in early stages.
And then there's Dispatch, which emerged from stealth in April 2026 with a $500,000 pre-seed round and an unusually ambitious thesis. The San Francisco startup, founded by ex-Astranis engineers Payton Case and Andrew Mello, is building refurbishable reentry vehicles and, longer-term, uncrewed "lights-out" space stations designed to power and service those vehicles. The roadmap calls for a 30 kg sub-scale demo in 2027, scaling to 300 kg vehicles supported by a 100 kW-class autonomous station by late 2029.
Case, in an interview with the industry publication Payload, framed the strategy bluntly: current government-lab approaches are "not commercially viable" for manufacturing at scale, and crew-rated destinations add cost without value for cargo-only industrial processes. Dispatch's pitch is to blend the flexibility of free-flying capsules with the power density of a fixed platform, minus the overhead of keeping humans alive.
Whether these timelines hold is anyone's guess. Dispatch has published a preliminary payload user's guide for its Free Flyer 1—a launch-agnostic reentry system with a 50×30×60 cm envelope, up to 180 days on-orbit, and recovery within six hours of landing—but the company has conducted a low-cost full-scale heat shield test while its orbital demonstration remains ahead. The guide carries a caveat: "pre-release" and "subject to change." That's a lot of caveats for a company with two listed team members on its Y Combinator profile as of the Spring 2026 batch.
The Science, Such As It Is
The technical case for microgravity manufacturing rests on decades of research, much of it conducted aboard the International Space Station. A meta-analysis published in Nature in 2024 cataloged experiments dating back to the Space Shuttle era, finding that semiconductor crystals grown in microgravity can exhibit improved structural uniformity, larger sizes, and better performance compared to terrestrial controls. The absence of buoyancy-driven convection and sedimentation allows for more controlled growth conditions.
But a November 2025 paper in Acta Astronautica injected a dose of caution. Simply "miniaturizing Earth processes" without re-engineering for the microgravity environment, the authors argued, can lead to disappointing results. The lesson from early ZBLAN optical fiber trials looms large: fibers produced in space showed promise, but required rigorous benchmarking against terrestrial standards before anyone could claim they were actually better.
Still, the proof points are multiplying. NASA and the ISS National Laboratory reported in November 2024 that Flawless Photonics had produced more than 7.4 miles of ZBLAN fiber during campaigns in early 2024, with visual comparisons suggesting reduced defects versus ground-manufactured fiber, though independent commentary in 2026 noted the need for standardized quality assessments before the industry can move beyond proof-of-concept. LambdaVision, working on a retinal implant, completed nine ISS missions by early 2024 and reached a 200-layer manufacturing milestone using automated layer-by-layer deposition. In December 2024, Space Tango and Auxilium announced what they called the first bioprinted drug-delivery device manufactured in microgravity on the ISS, part of an InSPA-funded initiative.
The pharmaceutical demonstration that's drawn the most attention—and the most venture capital interest—remains Varda and Improved Pharma's ritonavir work. Ritonavir, an HIV protease inhibitor, exists in multiple polymorphic forms. Form III is more soluble and potentially more effective, but it's metastable and difficult to produce at scale on Earth. In orbit, microgravity enabled the generation of Form III crystals, which survived reentry and returned to the ground intact.
Results posted in March 2024 confirmed the experiment worked flawlessly. A follow-on peer-reviewed study in npj Microgravity in April 2026 detailed the stability and survivability of the space-processed form. It's a narrow use case—an advanced drug formulation that might command premium pricing—but it illustrates the value proposition cleanly. Some materials are difficult or impossible to make on Earth. In space, with gravity out of the equation, they become feasible.
The semiconductor angle is more speculative but potentially larger in addressable market. Space Forge's December 2025 plasma milestone and the broader body of research on crystal growth in microgravity suggest that high-purity substrates for chips could benefit from orbital production. Yet McKinsey's April 2024 analysis noted that large-scale semiconductor in-space manufacturing is likely "10+ years" away from reaching scale economics, a timeline that feels both conservative and optimistic depending on which way the regulatory winds blow.
The CHIPS and Science Act, signed into law in August 2022 with $52.7 billion in appropriations for U.S. semiconductor manufacturing and R&D, creates policy tailwinds for onshoring advanced materials. Whether regulators and industry will treat space-sourced substrates as "onshored" remains an open—and politically fraught—question.
What Works, What Doesn't, What Might
Varda's operational tempo offers the clearest window into what industrial-scale in-space manufacturing might actually look like. The company has demonstrated that reentry licensing can move from a one-off approval to a vehicle operator license covering multiple missions. The W-series missions have doubled as hypersonic testbeds for the Air Force Research Laboratory and NASA, capturing data during reentry that is otherwise expensive to obtain. The expanded FAA license in June 2025 and the 20-reentry agreement with Southern Launch through 2028 suggest a pathway to regularized cadence.
Whether the unit economics pencil out is harder to assess from the outside. Varda hasn't disclosed per-mission costs or customer pricing. But the company is positioning itself as the lowest-cost option for rapid recovery of materials and as a platform for government science and technology work—a hedge, perhaps, if commercial pharma customers prove slower to materialize than hoped.
Space Forge is betting on a different horse: reusable heat shields and dedicated orbital furnaces. The Pridwen deployable heat shield completed zero-gravity deployment tests in 2025, aiming to eliminate the need for a fresh ablative shield on each mission. If the reusability thesis holds, Space Forge could drive down per-mission costs significantly. The ForgeStar-1 plasma furnace is already operational, and the company has signaled that future missions will move from proving the environment to returning actual semiconductor products.
The partnership with Intuitive Machines, announced in mid-July 2025, is designed to provide U.S. return capability, positioning Space Forge to serve American semiconductor manufacturers concerned about supply chain security—a selling point that may resonate in an era of heightened geopolitical tension over chip production.
Dispatch's approach, if it executes, would represent a third model: a network of return vehicles serviced by autonomous power-dense stations. The rationale is straightforward. Many in-space manufacturing processes—crystal growth, certain biotech applications—require sustained power and thermal control but not human presence. Commercial LEO destinations like Axiom's planned station or the NASA-backed Commercial LEO Destination projects (Blue Origin's Orbital Reef, Starlab) are designed to host astronauts, which drives up cost and complexity.
Dispatch's thesis is that a cargo-only, "lights-out" facility can strip out crew-rated systems and focus purely on supporting payloads and return logistics. The 100 kW target for the first station would exceed the ISS's payload power by a meaningful margin while avoiding the expense of life support.
It's an elegant concept on paper. But it depends on flawless execution of two hard things: a reliable reentry vehicle and an autonomous station. Dispatch has announced plans for a mid-2027 first flight and has completed a full-scale heat shield test, but an orbital demonstration remains ahead. And that $500,000 pre-seed is a modest sum to build orbital infrastructure. For comparison, Varda raised significantly more before its first flight, and Space Forge secured £22.6 million before ForgeStar-1 launched.
The Regulatory Piece (Which Matters More Than You'd Think)
The near-term trajectory hinges as much on regulatory momentum as on engineering breakthroughs. On that front, the signs are cautiously positive—though perhaps more in some jurisdictions than others.
The FAA issued a policy statement in April 2026 highlighting efforts to streamline commercial space license approvals, which matters as reentry cadence increases. But it's the UK that's moved most aggressively. On March 5, 2026, the UK government published joint guidance from the UK Space Agency, the Medicines and Healthcare products Regulatory Agency (MHRA), the Regulatory Innovation Office, and the Civil Aviation Authority. The guidance outlines an end-to-end pathway for space-manufactured medicines, including sandboxes and case studies to clarify how products move from orbit to patient access.
It's a world-leading step, and one that may pressure U.S. regulators to develop a parallel framework. As of mid-2026, the FDA's current Good Manufacturing Practice (cGMP) regulations still apply to space-made drugs without a dedicated space-manufacture pathway—a regulatory gap that could slow commercialization or push companies to seek approvals abroad first.
The ISS transition adds urgency. NASA selected SpaceX in June 2024 to provide the U.S. Deorbit Vehicle for the station's controlled reentry after 2030, a contract valued at up to $843 million. As the ISS winds down, the ecosystem of materials research and manufacturing demonstrations currently hosted aboard will need new homes. Some will migrate to commercial LEO destinations; others may move to free-flying platforms like Space Forge's satellites or, eventually, Dispatch's stations.
The timing is tight. Commercial LEO destinations are still in development, and Dispatch's 2029 station target leaves little margin for delay if it wants to capture ISS-adjacent demand. For that matter, it's not clear how much of that demand is real versus speculative—a question that applies to the entire sector.
The Trillion-Dollar Question: Will Anyone Pay?
Here's where the optimism meets the spreadsheet. In-space manufacturing market forecasts vary wildly—baselines published in 2025 range from $1.48 billion to $6.3 billion depending on methodology, and compound annual growth rate estimates span 8% to over 20% through 2032, according to reports from Precedence Research, 360iResearch, and others. These projections are, to put it charitably, indicative at best. Market sizing for industries that barely exist is more art than science.
What matters more is whether pharmaceutical companies will commit to multi-mission contracts for high-value drug formulations, whether semiconductor fabs will integrate space-grown substrates into supply chains, and whether optical fiber or advanced materials buyers will accept premium pricing for microgravity-produced goods.
Early signals are mixed. The ritonavir demonstration is scientifically impressive, but Improved Pharma has not disclosed commercial production plans or pricing. Flawless Photonics' 7.4 miles of ZBLAN fiber generated headlines, yet independent commentary noted the need for standardized quality assessments before the industry can move beyond proof-of-concept. LambdaVision's retinal implant work continues, but the path to regulatory approval and patient use remains long and uncertain.
Perhaps the sector's greatest advantage is that it doesn't need to replace terrestrial manufacturing—only to carve out niches where microgravity offers a decisive edge. A handful of ultra-premium pharmaceuticals. A few specialized semiconductor substrates. Some advanced optics. If the unit economics work for these verticals, the infrastructure being built today could scale. If not, well, the reentry vehicles and orbital platforms may end up serving adjacent markets—hypersonic testing, technology demonstration, even rapid return of time-sensitive cargo—before finding their original use case.
It wouldn't be the first time a space industry pivoted to survive.
What Happens Next
For founders, investors, and corporate R&D teams watching this space, the next 18 months will clarify which architectures have staying power. Varda's operational cadence, Space Forge's furnace-to-return integration, and Dispatch's station-plus-vehicle concept represent three distinct bets on how to industrialize microgravity. One or more may succeed. Others may pivot or fold. That's how infrastructure buildouts work.
What's certain is that the infrastructure layer is no longer speculative in the way it was even three years ago. The factories are going up, or at least the prototypes of factories. Capsules are landing in the Australian outback. Furnaces are firing in orbit. Regulatory frameworks are emerging, slowly, unevenly, but emerging nonetheless.
The question—the only question that really matters—is what those factories will produce, and whether anyone on Earth will pay enough for it to justify the extraordinary cost and complexity of building an industrial economy 250 miles above our heads.
The answer, like the capsules themselves, is still in flight.
