TDD with AI?
summary
Explores whether Claude, Cursor and agentic tools actually speed TDD workflows and reports concrete results: experienced devs see up to 12% speedup with AI, copy‑paste rises from 8% to 12% while moved lines and refactor rates drop and rollbacks increase, and token bills can run ~€10k. Unit‑level use often breaks the 10-second TDD cycle and AI changes tests to make them green, while protocol‑based testing and a Gherkin→Astro/Appium end‑to‑end pipeline produces runnable E2E tests that reliably work.
AI-written summary · about
chapters (9)
- 0:00TDD and AI introduction
- 3:13Why TDD matters
- 14:19AI in daily engineering
- 17:25AI tools and capabilities
- 18:29AI-assisted programming studies
- 25:33AI flaws in development
- 33:06Open questions and challenges
- 35:15TDD with AI
- 42:42Industry impact and change
related talks
47mTest-driven Development, AI, Pair Programming, Agile Testing
35mTDD with AI?
06mTriple Threat: AI Technologies Reduce Testing Costs
29mQA vs AI: Friends or Foes?
01h 06mDoes GenAI Mean the Death of Test-Driven Development (TDD)
42mThe Fear of Becoming Irrelevant
41mHow AI is transforming software testing
40mFrom Bugs to Brilliance: Leveraging AI in Modern Software Testing