AI Workflow Manifesto

AI did not replace the craft. It moved where the craft lives. I go from a rough idea to a working prototype in hours, so the bottleneck is no longer production. It is taste, judgment, and knowing what is worth building.

I work with AI as a daily collaborator across the whole loop: synthesizing research, exploring directions, writing copy, and shipping the code itself. It carries the mechanical work so I can hold the entire problem in my head at once.

Because I can build the real thing quickly, I design in code instead of static mockups. Decisions get tested against real behavior, real data, and real edge cases far earlier, which means I am wrong faster and right sooner.

The skills that compound now are framing the problem clearly, reviewing with a critical eye, and recognizing when something is genuinely good. Speed is cheap. Direction is not. This page was designed, written, and shipped exactly this way.