What this site is

This site is where I publish what I'm learning while building agentic AI systems for regulated environments.

The plan is one essay a week. Each one takes a single architecture decision and works through it, from the point of view of the person who has to get the system past security review and the risk committee. That means guardrails, approvals, agent identity, audit trails and the decisions behind them. From each essay I cut a carousel and a few short videos. The essay is the source and the rest is derived from it.

This site is the canonical home for all of it. The essays also go out by newsletter, so you can read them in your inbox if you prefer, but the version here is the one I keep up to date and the one I'll correct when I get something wrong. Videos and talks get their own pages as they're published.

The site itself is small on purpose. It's built with Next.js, the writing lives as MDX files in a git repository, and it's deployed on Vercel. There's no analytics, and the pages are static, so they should be quick on a phone.

One more thing, because it matters in this field: nothing here speaks for any employer, past or present. The opinions are mine.

If you want to get in touch, the about page has an email address.