Workshop Finds TDD Improves AI Coding Outcomes
A group of experts hosted by Martin Fowler and Thoughtworks met 25 years after the Agile Manifesto to assess AI's impact on software development, producing a Chatham House Rule report. The workshop concluded that test-driven development (TDD) yields dramatically better results for AI coding agents and shifts engineering rigor toward tests, architecture reviews and security. The report flagged unresolved issues around standardization and trust in non-deterministic systems.
Key Points
- 1Finds TDD yields dramatically better results from AI coding agents, preventing tests that confirm broken behavior
- 2Highlights shift of engineering rigor from coding to tests, architecture reviews and project decision points
- 3Recommends practitioners adopt TDD, strengthen architecture reviews and prioritize security for AI-assisted projects
Scoring Rationale
Actionable consensus on TDD and engineering discipline raises impact; limited novelty beyond a workshop constrains transformative significance.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
