Automate Chemistry.
Accelerate Discovery.
AutoRxn is a modern platform for reaction automation and scientific data orchestration using Bayesian Optimisation.
|
Why AutoRxn?
Reaction optimization is time-consuming and expensive. AutoRxn empowers chemists with intelligent tools to design, model, and refine experiments faster — grounded in real scientific data.
AI-Guided Experimentation
Bayesian optimization suggests new conditions from your data to explore chemical space efficiently.
AutoRxn proposes candidates that balance exploitation and exploration. You control bounds, constraints, and objective direction (maximize yield, minimize impurity, etc.).
Scientifically Grounded
Built by chemists for chemists — parameters, yields, and outcomes remain transparent and reproducible.
We keep full metadata for every run (reagents, temperatures, times, catalyst loadings). No black boxes — every suggestion is traceable back to prior evidence.
Data-Driven Decisions
Visualize relationships, spot outliers, and surface best-performing reactions in seconds.
Response plots and residual checks help you understand sensitivity to factors and detect data quality issues before they derail optimization.
Accelerate Discovery
Replace trial-and-error with predictive modeling that improves as you run more reactions.
Loop faster: import existing data, model the response, run the suggested batch, and watch the model confidence tighten with each cycle.
See the optimiser learn, point by point
Each new experiment updates the model, shrinks uncertainty, and pulls the prediction toward the true (unknown) function.
Use cases
Real workflows AutoRxn accelerates across chemistry and materials.
How AutoRxn helps: Catalyst Screening & DOEs
Prioritize catalysts, ligands, and bases with Bayesian optimization under tight screening budgets.