Designing a drug in the morning, reading lab data at night

Capable raises $12M from 50 Years, HOF and OpenAI executives to accelerate its Capable Core platform, designed to automate and speed up drug development.

Capable is emerging from its relative discretion. This young San Francisco company, founded by a team from MIT, Harvard Medical School, Roche, ETH, and Dana-Farber, presents itself as a lab capable of moving from AI-driven drug design to initial experimental data in twenty-four hours.

Its CEO, Isaak Freeman, bases his pitch on an observation: as models become more intelligent, the bottleneck shifts to the physical world, hence the idea of a "biological compute layer" that quickly produces data to feed the models. The company claims to have verticalized several steps of preclinical development into an in-house platform, the Capable Core, made up of six semi-autonomous modules rewritten to run, according to them, ten to twenty times faster.

The first stated target is unusual: peptides designed to reduce the need for sleep, with a focus on enhancing human capabilities rather than treating a disease. The stated goal is to have models iterate on hundreds of candidates to generate thousands of validated data points, and to reduce the search for therapeutic leads from several years to a few weeks. The company specifies that it doubles its key validations on cells and mice via external subcontractors, even if it is slower and more expensive.

Regarding funding, Capable announces a twelve-million-dollar fundraising round, led by 50 Years and HOF, as well as executives from OpenAI, including Wojciech Zaremba, Kevin Weil, Srinivas Narayanan, and Fidji Simo. At this stage, most of the public information relies on the company's own announcements, with the highlighted results not yet having undergone independent validation.