A staff engineer figuring out AI-assisted development in public — the tangents, the failures, and the learning.
This blog exists in the spirit of Austin Kleon’s Show Your Work. You won’t find polished tutorials here. Instead: scope creep documented in real-time, refactors that went sideways, a game that probably should have been paper-prototyped first, and the slow realisation that AI changes how you work but not how hard it is.
Start here
- The origin story — A Blog of Dubious Intent — why this exists and what you’ll find
- The game — Finally… A Wild MVP Appears — three months of building a tactical wargame with AI
- The katas — How I Learnt to Stop Worrying and Love Agentic Katas — structured practice for AI-assisted development
- The craft — The Smell of Panic When You Context Thrash — what goes wrong when you skip understanding
Latest
I Didn't Grok Superpowers
You Can’t Just Install Someone Else’s Workflow and Level Up.
The skills moment is happening. Matt Pollock’s repo is blowing up. Superpowers, nwave.ai. Curated bundles of markdown being treated as installable expertise.
Recently
The Best Part Has No AI in It
On building the plumbing between the prompts.
A friend of mine just got a 3D printer. He read my last post, the one about building a Soviet army in three evenings, and quite reasonably wanted to know how to do it himself. We were on our phones. He could not really sit and read the prompt in full, could not digest the style-versus-pose distinction, could not work through the order of operations. He got the idea but not the practice, and so he was rediscovering most of it from scratch.
An Army of One Prompt
On discovering that good process survives the jump from code to plastic.
That is how long it took to go from nothing to a printed 1000 point Soviet army for Bolt Action. Roughly fifty models, around forty of them unique sculpts, fully designed, generated, modelled, sliced, printed, and sitting on my desk. Concept to physical thing in my hand, three evenings, in my spare time, while doing everything else I usually do.
Show Your Work
On discovering that “show your work” is not the same thing as “do the work well.”
When I first started using Claude Code last summer, it felt like a chatty junior engineer who couldn’t wait to tell you what it was thinking. It kept you in the loop. It explained itself. It told you what it was about to try, why it thought that might work, and then narrated its way through whether it did. There was charm in it. You felt like you were pairing.