17:45 · Shift notes
Written down before I forget.
The blog. Notes from the stations: what worked, what broke and what I would try next.
- I drew my whole website in PythonMy old site said what I do and showed none of it. This one is a single isometric room, drawn by a Python script, coloured by CSS variables, and walked through over one working day.Read the note
- Building a software factory that learns from its own mistakesHow I combined two habits, agentic coding and a template for the parts every app needs, into a factory: what the first version got wrong, what I changed, and how every mistake now goes back into the template.Read the note
- Robots don't fail aloneI built a browser simulation of autonomous robots moving pallets around a logistics hall, with real vehicle numbers and a small business layer on top. Here is what it taught me about systems.Read the note
- Your agent will cheat on its testsWhat I learned building Agentest, an open-source test runner for AI agents.Read the note
- Building OpenFoliaI built the documentation system I wanted to buy. OpenFolia stores operating manuals as small modules in plain files that the manufacturer owns, then builds a correct manual for every product variant and every language.Read the note
- Freit: letting AI into a haulage office without handing it the keysFreit does the work of a transport office: it reads orders from mail, PDFs, scans and spreadsheets, sorts the inbox, proposes decisions and plans which truck takes what. A person makes every call that matters.Read the note
- What we learned building a small-model agent systemA 400M-parameter model, fine-tuned on about 1,100 real messages, came within 1.6 points of an 80B general-purpose LLM on a production routing decision. With a confidence gate that hands the unsure cases back to the LLM, it matched the LLM.Read the note