Clinical AI for the real world.

I’m a medical doctor with a PhD from Cambridge, working as a founding-style engineer with end-to-end ownership of clinical AI products used at national scale.

Patient access

The digital front door.

I owned and built major parts of Anima’s Requests product, specifically its dynamic questionnaire engine and patient experience, now used to support access for more than 10 million people.

Ownership: Requests and the products that grew from it.
Reach: More than 10 million people.

Clinical prioritisation

The right care, faster.

I built Smart Triage, turning patient requests into clinically safe urgency and routing signals.

Prioritise: Urgent patients surface first.
Route: Each request reaches the right team.
Outcome: In one ICB rollout, 80.4% were actioned the same day and 72.1% resolved without a GP.

Document automation

Information becomes action.

I was one of four engineers who built Anima Documents, including patient matching, clinical summarisation, coding suggestions, and follow-up workflows. It processes millions of documents across 25% of UK GP practices.

Scale: Millions of clinical documents.
Reach: 25% of UK GP practices.
Action: Patient matching, summaries, coding, and tasks.

Clinical infrastructure

Built into the workflow.

I built Anima’s EMIS Web and SystmOne integrations by myself, connecting Requests, Smart Triage, and Documents to the patient record.

Research foundations

From cells to systems.

My work spans biomedical machine learning, genomics, and clinical trials—from analysing genomic data for Nature Methods to platelet research in Blood and clinical-trial research in Nature Communications.

Publish: Nature Methods, Nature Communications, and Blood.
Author: Books on machine learning in healthcare for Elsevier.
Present: A billion-compound AI model at NVIDIA GTC.

Across the whole stack

See my work.

Read the story behind the work, browse my writing, or go straight to the evidence.

View GitHub