Built for the way chemists actually run experiments.
AutoRxn is a modern platform for reaction optimisation. It helps chemists design smarter campaigns, understand their data, and reach better conditions with fewer experiments – without having to become a data scientist.
We combine Bayesian optimisation, a chemistry-aware data model, and a clean interface designed around real lab workflows: substrates, catalysts, plates, vials, and instruments.
Make advanced optimisation as routine as running a TLC: fast, reliable, and accessible to every chemistry team.
AutoRxn closes the loop between experiment design, execution, and analysis for reaction chemistry.
Who we build for
AutoRxn is designed for teams that live at the interface of chemistry, automation, and data – from academic labs to scale-ups.
Academic groups
Method development, catalysis, and reaction discovery groups who want to get more from each campaign and train students on data-driven optimisation.
Startups & scale-ups
Chemistry-centred companies who need to move quickly, capture know-how, and make better use of high-throughput or flow equipment.
Industrial R&D teams
Process and discovery chemists looking to formalise optimisation workflows and align experimental design with business and scale-up constraints.
Principles we design around
Good optimisation tools shouldn’t feel like black boxes. AutoRxn is built around a few simple ideas that keep scientists in control.
Interfaces that look like your lab notebooks and plate maps, not generic dashboards. Chemistry concepts are first-class citizens.
Visual diagnostics, uncertainty, and clear assumptions so you can trust – and challenge – what the optimiser is suggesting.
Your data belongs to you. AutoRxn is built with privacy and security in mind, and we're opinionated about how scientific data should be handled.
AutoRxn is being built with experimental chemists, for experimental chemists.
We're actively working with early users to shape the product. If you're running reaction campaigns and want to see how AutoRxn fits your workflow, we'd love to hear from you.