Augmented Coding: Beyond the Vibes

How to use AI coding assistants for production-ready code, not just quick fixes

Augmented coding—where engineers direct AI to write quality, tested code—differs fundamentally from 'vibe coding,' where you accept whatever the AI generates. Kent Beck built a production-competitive B+ Tree library in Rust and Python by treating the AI as a senior pair programmer: enforcing TDD discipline, catching design drift early, and refusing to accept shortcuts like disabled tests or unasked-for features. The result reshapes what programming becomes when you offload syntax and boilerplate but retain architectural judgment.

Read full essay on Substack ↗

Questions this essay answers

  • What's the difference between using AI to ship fast versus using AI to write good code?
  • How do you keep an AI coding assistant from adding complexity you didn't ask for?
  • Can AI write production-ready, performance-competitive library code if you enforce TDD?
← All essays