Taming the Genie: "Like Kent Beck"

How do AI coding assistants adopt better design practices through prompting and constraints?

Personas improve micro-behaviors like testing style and naming, while architectural constraints shape macro-decisions like class hierarchies. The combination—explicit persona plus design guidance—produces the best results, but the real leverage comes from running millions of design contests to evolve coding practices through selection pressure rather than encoding human expertise.

Read full essay on Substack ↗

Questions this essay answers

  • Does asking an AI to 'code like [expert]' actually improve code quality, and how?
  • Should I guide AI assistants through personas or through explicit architectural constraints?
  • How do you evolve AI coding behavior beyond pattern-matching to real design judgment?
← All essays