Warp Sciences has raised $215 million to build a U.S.-based chemistry contract research organization that combines chemists, robotics and AI to make and test molecules for drug discovery teams. The South San Francisco company has come out of stealth with Lightspeed leading the round.
The company describes its chemistry CRO as infrastructure for biotech and pharmaceutical firms that want to find new medicines faster. It says it has no proprietary drug pipeline of its own.
What Warp is building
Warp's first service is lead optimization analog synthesis. In that work, chemists make variations of a molecule so scientists can compare them while searching for a drug candidate. Experienced chemists will guide the experiments at the start, and the lab will become more automated through robotics and AI over time.
The stated goal is to generate experimental data that trains better AI models, which in turn support more capable labs. Warp says it is designing automated labs that can run experiments, learn from the results and improve with each cycle.
Chief executive Kevin Weil framed the problem as one of physical data. "AI has dramatically accelerated software engineering and mathematics, but it has yet to make the same impact in the physical sciences," he said. He added that Warp is starting by making the research compounds drug discovery needs today while it builds the labs and automation to shorten the cycle from hypothesis to experiment to answer.
Chief technology officer Manohar Paluri described the intended loop: design experiments, execute them, and learn from the outcomes. He said the company is building the software, AI and robotics alongside the lab so each part is shaped by what chemists need.
The sovereignty argument
Warp is also selling on location. The company points to a 2024 survey by the Biotechnology Innovation Organization in which 74% of respondents reported contracts with Chinese companies for preclinical and clinical services. The BIOSECURE Act, signed into law in December 2025, set up a framework for restrictions on federal procurement and funding involving designated biotechnology companies of concern.
Warp says it is positioning itself as a U.S. chemistry partner that offers speed, quality and close collaboration, and that gives customers more control over where sensitive research takes place.
Chief scientific officer Solomon Reisberg, previously in medicinal chemistry leadership at Terremoto Biosciences, said he has worked with overseas chemistry partners that delivered speed and value but left him concerned about protecting program intellectual property. He said Warp aims to offer the same caliber of service in the U.S. with IP protection at the center, and without the price premium that U.S. boutique CROs currently charge.
Team and investors
Weil was most recently chief product officer and vice president of science at OpenAI. Paluri was a vice president at Meta Superintelligence Labs and a founding member of FAIR. DJ Patil was a founding executive at Devoted Health and the first U.S. chief data scientist for the White House. Early hires bring robotics experience from Genesis, Skild AI and 1X Technologies, along with scientific experience from Genentech and Lawrence Livermore National Labs.
Patil said customers decide which molecules to pursue and keep control of their drug programs. "We're not developing drugs of our own," he said. "We're building the infrastructure for the people who do."
Beyond Lightspeed, participants in the round include Premji Invest, Mayfield, IVP, Lux, Conviction, Menlo Ventures, 01 Advisors, Standard, Lowercarbon Capital, Gigascale Capital, Angular Ventures, Scribble Ventures, Uncork Capital, Innovation Endeavors, Matter VP, Zero Shot Fund, SignalFire and Moxxie Ventures. Angel investors include Brian Armstrong of Coinbase, Mike Krieger, Milan Kovac, Ion Stoica, Fidji Simo, Brendan Foody, Mark Chen, Sarah Friar, Deepak Pathak and Lachy Groom, among others.
What this means and what is unproven
The structure of the business is the notable part. Warp sells a service, not a drug or a robot. Customers pay for compounds, and the lab collects the experimental data as a by-product. If that works, revenue from a conventional CRO funds the build-out of automation, and the data feeds models that make the next cycle faster. That is a different route from companies that build a lab robot and sell it, or from drug developers that automate their own pipelines.
The funding size and the investor list show that capital is flowing toward physical-world data for AI. Weil's argument rests on a real gap: language models improve on abundant text and code, while chemistry needs molecules to be made and measured. A CRO is a practical way to get a steady supply of that work from paying customers.
The sovereignty pitch adds a second driver. The survey figure and the BIOSECURE framework give U.S. companies a reason to consider domestic chemistry capacity. Reisberg's comment about price is the harder test, since Warp says it will avoid the premium of U.S. boutique CROs while offshore partners are valued for speed and value.
Several points remain unproven. The source gives no timeline for automation, no throughput or turnaround figures, no pricing, no customer names and no description of which robotic systems will run in the lab. Early work depends on experienced chemists, so the shift to largely automated synthesis is a plan, not a result. It is also unstated how much data the lab will generate, or whether models trained on it will improve compound design in a way customers can measure.
The practical measure will be simple. Watch whether Warp delivers analogs faster or cheaper than existing providers, and whether it can show that each cycle improves the next. Until then, the company has money, a team and a thesis, and not yet evidence that closing the loop works in a commercial chemistry lab.