Human-AI collaboration
My DPhil research studies how people and AI systems can work together safely and effectively in healthcare, with responsibility allocated to the right decision-maker.
About
I am a DPhil student in Engineering Science at the University of Oxford, working on trustworthy artificial intelligence for healthcare. My research focuses on how people and AI systems can collaborate safely and effectively in clinical decision-making.
I develop learning-to-defer systems that recognise when human expertise is needed, including guided deferral with language models and deferral to previously unseen experts.
Before starting my DPhil, I worked as a data scientist across NHS England and an NHS trust. At NHS England, I contributed to a Bayesian hierarchical early warning system for national COVID metrics and worked on machine learning products from large-scale data preparation through production deployment.
Research
Learning to defer
I develop machine learning systems that recognise when human expertise is needed, including guided deferral with language models and deferral to previously unseen experts.
Applied clinical data science
My work is grounded in practical healthcare settings, including experience working on national-scale forecasting systems and machine learning deployment at NHS England.
Selected publications
CV
Experience
Data Scientist
NHS England
Worked on a Bayesian hierarchical early warning system for national COVID forecasting, contributed to machine learning products from data preparation to deployment, and mentored junior analysts and MSc researchers.
Education
DPhil in Engineering Science
University of Oxford
Thesis: Human AI Collaboration in Healthcare. Supervised by Professor Alison Noble CBE FRS FREng FIET and Professor Helen Higham. Fully funded EPSRC Studentship.
MSc in Data Science, Distinction
University College London
BSc (Hons) in Mathematics, First Class
University of York
Teaching & service
Graduate Teaching Assistant for group-based AI projects on doctoral training programmes and international visits.
Reviewer for NeurIPS 2024, ICML 2024, and ICLR 2025.
Contact
Please feel free to get in touch via LinkedIn.