About

I am a doctor, health data scientist and applied AI engineer. I trained in medicine, practised as an anaesthetist in the NHS for more than ten years, and am completing a PhD in health data science and machine learning at the MRC Integrative Epidemiology Unit, University of Bristol.

My work sits where clinical medicine, statistical genomics, machine learning and AI engineering meet. In research, I ask why some patients develop serious complications after major surgery, whether we can predict them, and which biological drivers are genuinely causal. In industry, I build the AI systems that make data and evidence usable by the people who need them.

Research Focus

  • Statistical genomics & causal inference: Genome-wide association studies, polygenic risk scores, genetic correlation and Mendelian randomisation in UK Biobank, to map the genetic architecture of postoperative delirium and other complications.
  • Multiomic machine learning: Reproducible pipelines integrating proteomic, metabolomic, biomarker and clinical data in cohorts of up to 150,000 people, evaluated with nested cross-validation and decision-curve analysis to test whether added biology improves real decisions.
  • Applied AI & agent engineering: Designing and shipping production multi-agent LLM systems, from orchestration and tool calling to the evaluation benchmarks and human review that make them trustworthy.
  • Open science: Analysis code, pipelines and GWAS summary statistics are released publicly alongside each paper.