Kevin Weil launched Warp Sciences to combine AI, robots and human chemists in drug discovery.
Why it matters: Warp is targeting the physical laboratory work that software-only scientific AI cannot perform. If the model succeeds, it could increase experimental throughput and change how drug developers organize chemistry work.
- Warp reportedly raised $215 million in a Series A led by Lightspeed Venture Partners.
- Bloomberg Law reported a roughly $865 million post-money valuation; the supplied materials identify Bloomberg Law as the sole source for that figure.
- Human chemists will wear cameras and sensor gloves to create training data for robots and AI models.
- Warp will initially focus on lead-optimization chemistry - making and testing molecular variants to improve a potential drug compound.
Kevin Weil, a former OpenAI executive who led a science-focused effort, announced Warp Sciences in a post on X. The supplied materials do not identify Weil's exact former OpenAI title or provide a separate source for that role.
Bloomberg Law reported that Warp raised $215 million in a Series A led by Lightspeed Venture Partners, with participation from IVP, Mayfield, Lux Capital, Premji Invest and Menlo Ventures. Bloomberg Law is the sole source in the supplied materials for the reported financing and approximately $865 million post-money valuation. Weil's X post provides the company-side announcement of the launch and financing.
Warp is building laboratory infrastructure and chemistry services for drug developers rather than developing its own medicines. Its initial focus is lead-optimization chemistry, the stage where researchers make and test different versions of a molecule to improve its properties.
Human chemists will perform much of the initial laboratory work while AI helps plan compound synthesis and select the next experiment. Cameras and sensor gloves will capture their movements and laboratory data, creating demonstrations for robots and models.
Warp says robots could handle repetitive, high-volume steps such as setting up reactions and purifying samples. The company also says experiments will generate chemistry, robotics and process-engineering data, including information from failed runs. Weil said, "Each experiment that we do will make the next one better."
The approach follows earlier autonomous-lab demonstrations. A University of Liverpool robot completed 688 experiments over eight days, operating for 172 of 192 hours, according to the University of Liverpool.
By the numbers
- $215 million - reported Series A financing
- $865 million - reported post-money valuation, according to Bloomberg Law
- 688 experiments - completed by a University of Liverpool robot over eight days
Yes, but: The financing, participating investors and valuation are supported in the supplied materials primarily by Bloomberg Law; the company-side X post is not an independent financing report. Warp has not disclosed an operational start date for its human-supervised chemistry work.
What's next: Warp has disclosed plans to begin with lead-optimization chemistry and human chemists supervising laboratory work, but it has not announced an operational start date.