The REWORK Podcast — selected coding-agent conversations
Building with LLMs, finding a co-founder & other listener questions
Seven episodes on agent-assisted development, durable software, product judgment, and agent-ready interfaces
A manually reviewed selection from The REWORK Podcast following 37signals’ move from cautious LLM assistance to terminal agents, AI-accelerated Basecamp development, agent-accessible product interfaces, and agent-built competition. Jason Fried and David Heinemeier Hansson emphasize disposable prototyping, senior review, architecture, product scope, and consequence-aware use. Their examples are first-party experience and forecasts, not controlled measurements.
In this collection Browse 7 summaries 7 of 7
Can a designer use Claude Code instead of a programming co-founder? In REWORK's opening listener question, David Heinemeier Hansson and Jason Fried argue that the answer depends on what can go wrong—and whether the builder can maintain the result. Independent iOS developer Ryan Chitwood poses the question to host Kimberly Rhodes.
Key Points Covered
- The use case depends on criticality: Hansson says agents can expand what independent builders create, but scale, fidelity, sensitive data, and the consequence of failure should determine the release standard 01:43–03:49(opens in a new tab).
- Low-stakes prototypes are a strong starting point: Builders can make ideas tangible, learn quickly, and discard approaches without exposing important data 03:49–04:51(opens in a new tab).
- Maintenance competence remains a boundary: Fried warns that a solo builder can move beyond their understanding and become unable to diagnose or repair the system when the model cannot recover 04:51–06:52(opens in a new tab).
Full video: https://www.youtube.com/watch?v=GifToS0FTII(opens in a new tab)