Three washing machine-sized capsules are circling Earth at roughly 300 miles altitude, each operated autonomously by California-based Varda Space Industries. One is growing crystals of a potential cancer drug in microgravity. The other two are testing re-entry thermal protection and navigation — the unsexy but load-bearing engineering that determines whether any of this becomes a business. Varda launched two capsules on 1 October and a third on 20 October, marking the first time any company has simultaneously operated multiple commercial re-entry vehicles in orbit. The physics case is real. Gravity causes particles to settle and density differences create convective currents, both of which introduce defects into crystal growth. Remove gravity and you get more ordered crystal structures. For pharmaceuticals, crystal geometry determines how a drug dissolves, its stability, and its delivery method — potentially converting IV-only drugs into injectable formulations. The same logic extends to semiconductors and exotic materials like graphene, where terrestrial manufacturing consistently produces defect-ridden crystals that underperform their theoretical potential. Decades of ISS experiments have validated the science at small scale, but the ISS was never a factory. It is a crewed research station — dirty, vibration-prone, expensive to operate, and scheduled for retirement around 2030. Varda's free-flyers solve this by removing humans entirely: autonomous labs hitch rides on SpaceX rockets, process materials in orbit, then separate a capsule that re-enters the atmosphere and parachutes to an uninhabited landing zone. The company plans seven more missions in 2027 and ten in 2028, aiming for the kind of iteration frequency that drug development timelines demand. Varda is not alone. UK-based BioOrbit recently sent a microwave-sized lab to the ISS targeting self-injected cancer drug production. Cardiff's Space Forge is pursuing a hybrid model for semiconductors: grow seed crystals in orbit, return them to Earth as templates for bulk terrestrial production — a sourdough-starter approach to materials science. Space Forge's ForgeStar-1 recently demonstrated some manufacturing processes but has not yet completed a full produce-and-return cycle, with that milestone targeted before 2030. The economic question remains wide open. Falling launch costs — driven primarily by SpaceX's reusable rockets — have made orbital manufacturing physically plausible, but nobody has yet demonstrated that the per-unit value of space-manufactured materials justifies the per-unit cost of getting them there and back. As Satellite Applications Catapult's Dr. Meganne Christian put it plainly: "It's not going to be commercially viable for every application. We have to identify those high-value use cases where it makes sense." Regulatory infrastructure is playing catch-up. The UK's MHRA is expected to publish guidance later this year on how existing medicines regulations apply to drugs manufactured in orbit — a necessary precondition for any space-made pharmaceutical to reach patients. The deeper regulatory challenge is consistency: proving that an autonomous orbital factory can produce results reliably enough to satisfy pharmaceutical quality standards. Crystal growth experiments need to work not once but hundreds of times. The trajectory here is genuinely generative — new manufacturing capabilities that terrestrial physics cannot replicate — but the gap between demonstration and commercial viability is enormous. Varda's real achievement so far is not the drug crystals; it is proving that autonomous capsules can launch, operate, and return repeatedly. That is infrastructure. Whether the economics ever close depends on whether high-value use cases (oncology drugs, specialty semiconductors) can absorb launch costs while delivering materials that are measurably superior to their terrestrial counterparts.
