Essays and field notes on the work that got more interesting, not less: ambitious systems, the problems AI creates as fast as it solves them, architecture, and the craft of shipping.
Domain Driven Design never failed as an idea. It failed as an enforcement mechanism. Here is what changes when the code is written by twenty agents instead of one person, and why architecture has to become something you can actually run.
A market research demo took me a few days to build and can stand in for weeks of manual work. Here is why I keep giving these tools away instead of selling them, and why the thinking behind them isn't going anywhere.
Working with many agents, git worktrees, and a wiki the agents maintain themselves.
No posts in that category yet, more soon.
If your team rebuilds the same class of study for every new market, I build the system that runs that assembly end to end. Start with a conversation, or watch a live study first.
