DHH’s new way of writing code

By Pragmatic Engineer

Categories: Product, Startup

Summary

DHH did a complete 180 on AI coding after dismissing it 6 months ago, now building everything AI-first and finding his team more ambitious than ever. Ruby on Rails is experiencing a renaissance due to token efficiency and readability for AI agent workflows.

Key Takeaways

  1. AI agents are enabling projects that would never have been greenlit before—37 Signals tackled P1 optimization (fastest 1% of requests) purely because AI made it feasible to attempt.
  2. Token efficiency matters significantly for AI workflows; Ruby on Rails became ideally suited for agent-based development because it produces readable, verifiable code with fewer tokens than alternatives.
  3. Effective AI leverage creates a tension: developers working with agents report working harder than ever despite increased productivity, requiring intentional boundaries on supervision time.
  4. Craft and aesthetics in code become more valuable in an AI era—beautiful, well-designed software is more likely to be correct and easier for AI agents to understand and modify.
  5. Personal fit and taste compound across markets; building something that perfectly solves your own problem attracts thousands of similar users, as proven with Rails, Kamal, and Umachi Linux.

Topics

Transcript Excerpt

I feel that you very much value software engineering as a craft >> hugely. I mean I think aesthetics is truth. When something is beautiful, it's likely to be correct. I think this is true in mathematics. This is true in physics. This is true in a lot of different domains. >> I wonder if there's a part of AI about the impact of doing work that we would have not done before. >> The number of projects we have tackled internally that we would never even have contemplated starting on or Legion. Jerem...