CellType is building agentic drug discovery on simulated human biology, an approach that lets AI agents test hypotheses in software before experiments move into the lab. The company models cells, tissues, and disease pathways as computational environments, giving agents a place to run thousands of in silico experiments and surface the most promising candidates for wet-lab follow-up. That distinction matters: it turns discovery from a bench-bound craft into something closer to an engineering discipline, where failures are cheap and iteration is continuous.
The pitch is speed and cost. Traditional drug discovery is measured in years and hundreds of millions of dollars, with most candidates failing before they reach a clinic. Simulated biology flips the early phase of that process: instead of a small team running a handful of assays per month, agents can explore broad chemical and biological spaces overnight, letting researchers concentrate bench time on the ideas most likely to survive. The result is a narrower, better-informed pipeline entering the expensive stages of development. None of this replaces the lab, but it changes what the lab is asked to do: verify, rather than discover, the most credible leads an AI agent has already stress-tested.
CellType joins Robinhood Ventures Fund II as part of its healthcare AI cohort. The fund counts more than 80 companies across its portfolio, backs founders with $250K SAFEs, and sources heavily from the Y Combinator network. It is set to begin trading on the NYSE under the ticker RVII on August 13. For CellType, the backing arrives as biology simulators and agentic systems converge into a genuinely new way to run early drug research, and the fund’s software-first thesis fits a company whose first experiments happen on a server rather than a bench.
