Updated June 2026 · Cardiff, Wales
Working on
- Preparing for a panel interview for a Clinical Safety Officer role — the role I've been working toward for the past 18 months. Panel is end of June.
- Building this site as a long-term deep work project. Slow, deliberate, owned. The alternative to a LinkedIn posting schedule.
- Writing the first piece: a reflection on what I learned from writing a self-authored clinical hazard log for an AI workflow.
- Continuing NHS clinical work in Tier 3 weight management while the broader pivot completes.
Reading
- Deep Work — Cal Newport. The reason this site exists rather than a LinkedIn posting schedule.
- DCB0160 Clinical Risk Management: its Application in the Deployment and Use of Health IT Systems — the NHS standard that governs clinical safety of health technology. Less readable than Newport.
- NHS AI Lab publications on AI governance and safety assurance in clinical settings.
Thinking about
- The gap between how AI safety is conceptualised in technology and how clinical risk governance actually works inside the NHS. They're speaking different languages and not many people are translating.
- What it means to build a professional reputation slowly and deliberately — through depth rather than volume. Whether that actually works in a field that currently rewards visible activity.
- The longer-term picture: what clinical safety governance looks like in ten years when AI is embedded rather than novel, and where the interesting problems will be.