This is the first post on the site, so a quick introduction felt like the right place to start.

I’m Thomas, an applied AI engineer based in London. Most of my work sits in the gap between research, engineering, and product: messy ideas and half-proven prototypes that need to become systems people can actually use.

I like building things that are clear, practical, and a little opinionated. Usually that means tools, workflows, and small systems that help people get real work done without adding more noise than they remove.

What I want to build

The ideas I keep returning to cluster around the same themes:

• AI tools that are genuinely useful rather than just impressive in a demo
• personal software that’s lightweight and easy to own
• publishing workflows that make it easier to write consistently
• systems that leave a clear record of how they work and how they changed

Some of that will show up here as code. Some of it as notes, experiments, dead ends, or things I’m still trying to understand properly.

This isn’t meant to be a polished highlight reel. I’m more interested in having a place to think in public, document progress honestly, and make the work legible over time.

Why this site exists

Social platforms are good at momentum, but bad at memory.

You can post an update, share a screenshot, mention that something shipped, and that’s useful in the moment. A few weeks later it’s buried. The context is gone, the tradeoffs are gone, and the thread of the work is hard to recover.

I wanted a place that works differently. Somewhere between a lab notebook, a project log, and a personal website: a home for ongoing projects, weekly notes, and longer reflections on what I’m building.

The plan is to keep a readable record of what changed, why it mattered, and what comes next. Not to publish essays every time I touch a repo.

What you’ll find here

Roughly:

• weekly notes on active work
• project pages with longer-running context
• occasional posts on design decisions and lessons learned

Some weeks will have more to show than others. Some posts will be short. Some will probably start as a way to clarify my own thinking before they’re useful to anyone else. That feels fine.

A small promise to myself

I want this site to be the clearest place to understand what I’m working on. If it works, it should get easier to keep building, easier for people to follow along, and easier to look back later and see something more coherent than a pile of disconnected updates.

Anyway, thanks for stopping by early.