01
Scientific reasoning systems
Building AI systems that help frame questions, organise evidence, and support the reasoning process behind scientific discovery.
I am based at the Cavendish Laboratory at the University of Cambridge, where I work on problems connecting modern AI methods with astrophysical and cosmological research. I am also associated with the Kavli Institute for Cosmology, Cambridge and part of the Infosys-Cambridge AI Centre.
My interests include scientific machine learning, the design of agentic systems that support or automate parts of the research process, and the application of AI to a broader range of cosmology research questions. More broadly, I am interested in using AI to accelerate scientific discovery while keeping methods robust, interpretable, and useful to researchers in practice. I am supervised by Boris Bolliet.
I began this line of work in my master's dissertation for the MPhil in Data Intensive Science at Cambridge, where I investigated parameter inference using diffusion models. That work continues to shape how I think about AI for scientific discovery.
Outside research, I enjoy sport, and you are likely to find me rowing on the River Cam.
