ML for early detection with syndromic surveillance
As part of my PhD, I have been working on a pre-syndromic surveillance tool which groups symptoms reported from emergency rooms into clusters of disease using ML. I’m doing this as a first step in improving early detection with real-time syndromic surveillance in the UK. My next step will be to work on anomaly detection algorithms.
Functional prediction for hazard screening
Between September 2025 and February 2026, I’m scoping out what a government-funded def/acc programme could look like for functional prediction. It’s inspired by IARPA’s FunGCAT programme and the subsequent rapid progression of biological language models. I am doing this as a contractor at the Centre for Long-term Resilience.
Building a customer screening tool for DNA providers
Alejandro Acelas and I have made Cliver, an AI-powered customer screening tool. It searches public databases, reviews publications, and checks affiliations to create a risk report of customers attempting to order synthetic DNA. If you would like to try it, please reach out through the website!
Writing Fermi/BOTEC questions for the Estimation Game
Every month I come up with a theme and questions for the Estimation Game run by Sage Futures. I’m biased of course, but I think this is a super fun way of improving your calibration and gaining some very specific knowledge in quite general topics. Previous themes included pandemics, AI progress, the history of war and midsummer. Sage Futures has a lot of other cool tools for forecasting and for explaining AI capabilities, check them out!
Crocheting
