Building the AI Virtual Cell to decipher biology and disease
The cell is the fundamental unit of life. Rivercell is building the platform to understand how cells change in disease and predict how they respond to treatment.
Cell response to treatment remains largely unpredictable
Treatment response depends on cellular state, genetic background, environment and previous exposures. Cells can also adapt and develop resistance.
We need to predict which interventions, alone or in sequence, restore function or eliminate diseased cells while limiting harm to healthy tissue.
Four recurring gaps in today's cellular data
Observational
Cellular states are recorded without testing responses to interventions.
Monomodal
A single readout leaves molecular and cellular responses unlinked.
Static
A single timepoint leaves each cell’s response over time unobserved.
Context-poor
Too few cell types, genetic backgrounds and environments limit prediction in new contexts.
The platform that generates the right data at scale to unlock the AI Virtual Cell
We are building a proprietary, programmable platform and a world model of the cell as one lab-in-the-loop. The model will choose experiments; the platform will run them at scale. Each result will improve predictions and guide the next experiment.
Biological inputs
Rivercell platform
World model of the cell
Perturbational
Interventions tested against controls, alone, in combinations and in sequence.
Multimodal
Live-cell imaging linked to endpoint gene expression in the same cells.
Longitudinal
The same cells tracked over time to distinguish transient responses, adaptation and delayed effects.
Context-diverse
The same interventions tested across cell types, genetic backgrounds and environments.
Join our team
We are hiring scientists and engineers in Paris across thin films, fluidics, biotechnology and AI.
See our open positions → Partnerships · WorldwidePartner with us
Partner with us on perturbation experiments, single-cell data and predictive models for drug discovery.
Get in touch →