Build-in-public notes, essays, and field reports from the practice. Mostly about shipping AI software end-to-end as one person.
A quiet pattern emerged: the projects that worked had one decision-maker, one outcome, and one deadline. Everything else is preamble.
Hourly billing rewards me for being slow. Fixed price rewards both of us for being specific. I'll choose specific every time.
The job most senior designers do in 2026 isn't designing software — it's coordinating the design of it. AI changed what I want to do for work. This is what re-energised me.
Most AI features I build take a week of model work and three weeks of interface design. The model is a component. The product is the work.
I turn down more projects than I take. The math says I'd have to. The interesting question isn't whether to say no — it's how to say it well, and to what.
An AI itinerary tool for travel agents. Sixteen months of building, the lessons that stuck, and the mistake that taught me everything I now bring to every AI build.
A deliberately small practice. Four fixed-price AI builds a year, no retainers, no consulting work without a build outcome. The volume cap isn't a feature of the marketing — it's a feature of the work.
I won't run the build/buy decision for you — that's strategy work I don't sell. But I've watched enough builds go right and wrong to have a strong view on the patterns.
AI doesn't replace a process — it changes the shape of one or two specific moments inside one. Here's where I keep finding the leverage, and where I'd urge caution.
The most common ask I get from operations leads is 'we know we should be automating things — where do we start?' The honest answer is 'smaller than you think, sooner than you think.'
The most common SaaS-architecture mistake I see is treating scalability as a future problem. 'We'll worry about scale when we have users' is a slow rewrite waiting to happen.
A roadmap pretends you know what's worth building before you've built anything. With AI, you don't. Plan a pilot, not a roadmap.
Twenty years of building software, and the last twelve months are the largest single shift I've seen. Not because AI is replacing the work — because it's changed the shape of every hour I spend at a keyboard.