Luisa Crawford
Jan 12, 2026 23:20
San Francisco startup uses NVIDIA Isaac Sim and Omniverse to automate cell therapy manufacturing, targeting $38B market with robotics that cut contamination risk.
Multiply Labs is betting that robots can do for cell therapy labs what automation did for semiconductor fabs—and the numbers suggest they might be right. The San Francisco startup claims its robotic manufacturing systems cut production costs by more than 70% while reducing contamination risk in one of medicine’s most finicky processes.
The company, founded in 2016 and backed by Y Combinator, now works with major cell therapy players including Kyverna Therapeutics and Legend Biotech. Its pitch is straightforward: cell therapies are personalized treatments that require extracting cells from patients, genetically modifying them, and reinfusing them to fight diseases. Each batch is essentially a one-off production run where a single contamination event can destroy weeks of work.
“She showed me what she did in a lab and how difficult it was, and I couldn’t believe it—I thought drugs were made like chips, and this was insane but also real,” said Fred Parietti, co-founder and CEO, recalling when MIT colleague Alice Melocchi first showed him the manual processes involved.
The Tech Stack Behind the Clean Room
Multiply Labs is building its automation platform on NVIDIA’s robotics infrastructure. The company uses NVIDIA Omniverse to create digital twins of lab environments and Isaac Sim to train robots on the precise movements required for cell handling. More recently, they’ve started developing humanoid robots using NVIDIA’s Isaac GR00T foundation model.
The simulation-first approach matters here. Cell therapy manufacturing involves what the industry calls “tacit knowledge”—the undocumented expertise that experienced scientists develop over years. Multiply Labs uses imitation learning to capture these skills by analyzing video of expert technicians, then translating those movements into robotic control policies.
“It needs to be sterile, and you don’t want anyone breathing anywhere near the cells, so it was an obvious high value application of robotics,” Parietti explained.
Market Timing and Regulatory Tailwinds
The cell therapy market stood at roughly $6.88 billion in 2025 and is projected to hit $38.24 billion by 2034, according to Straits Research. That growth trajectory depends heavily on solving the manufacturing bottleneck—current methods are expensive, inconsistent, and struggle to scale.
Regulatory momentum appears to be building as well. The FDA announced increased manufacturing flexibility for cell and gene therapies on January 12, 2026, potentially smoothing the path for automated production systems.
For investors tracking the intersection of AI, robotics, and biotech, Multiply Labs represents a pure-play bet on physical AI in healthcare manufacturing. The company remains private, but its partnerships with publicly traded cell therapy developers like Legend Biotech (LEGN) offer indirect exposure to this automation trend.
The real test comes as these robotic systems move from pilot programs to full-scale production. If Multiply Labs can consistently deliver on that 70% cost reduction at scale, the economics of personalized medicine start looking very different.
Image source: Shutterstock
