€22m · Seed · Healthtech · Paris, France
Rivercell, a Paris-based biotechnology and AI company building a system to predict how human cells respond to drugs and genetic changes, has raised a €22m Seed round. The company said HV led the financing, with HCVC, Alven and Bpifrance Digital Venture participating. The capital will scale Rivercell's proprietary data-generation platform, expand its automated wet lab in Paris and support the launch of its AI Virtual Cell programme.
The model depends on the experiments
Drug-discovery models need evidence about what happens after an intervention, not only snapshots of cells in an untreated state. Rivercell's published technical plan centres on perturbation experiments that expose cells to drugs, genetic changes or environmental shifts, then connect live-cell imaging with endpoint gene-expression measurements. The company intends to track responses over time and across different cellular contexts so its model can learn how state, sequence and timing change an outcome.
That makes Rivercell's wet lab part of the product rather than a validation service added after the model is trained. The model is meant to choose informative experiments; the platform runs them; and each result updates the next prediction. If that loop works as described, the company can build a proprietary dataset around the questions its model finds most uncertain, instead of depending only on public atlases that were created for other research goals.
The commercial implication is a demanding one: customers will judge the system by whether its predictions help narrow expensive laboratory work and produce hypotheses that survive experimental testing. More data alone will not establish that value. Rivercell must show that its paired hardware, assays and modelling workflow generates reproducible responses across cell types and treatments, and that those responses improve decisions in real drug programmes.
Experience across diagnostics and biological AI
Rivercell was started in summer 2025 by CEO Yann Fleureau, who previously co-founded AI diagnostics company Cardiologs, and was joined in 2026 by CSO Eric Durand. Rivercell's announcement says Durand previously led oncology data-science work at Novartis, served as Chief Data Science Officer at Owkin and co-founded Bioptimus. The company has also appointed EPFL researcher Charlotte Bunne and Helmholtz Munich's Fabian Theis to its scientific advisory board.
For a newly launched company, the Seed round therefore finances both infrastructure and proof: Rivercell needs to turn a technically ambitious experimental loop into evidence that pharmaceutical researchers can trust. The expansion of its Paris wet lab and launch of the AI Virtual Cell programme are the first concrete tests of whether that full-stack approach can shorten the path from a biological question to a validated result.
