BDD, ADR, PRD, WTF: Capturing Decisions for Humans and AI Alike
summary
Executable BDD scenarios tie directly to PRDs and critical user journeys so the claim is testable not just described; the team ships Cucumber features that reviewers read and run. ADRs record why decisions exist and failed agent commits link back to the document. Enforcement runs in git hooks and CI: module import linting prevents N+1 by forbidding DB access from templates and E2E suites, and a new tool Spec 27 tests agents that run 20 to 50 context compacts.
AI-written summary · about
chapters (9)
- 0:00Decision capture intro
- 1:53Why documenting decisions matters
- 2:57ADR for architectural choices
- 3:40PRD for product goals
- 4:40BDD specs
- 6:06Design systems with AI
- 7:28Git hooks and CI checks
- 9:37Feedback loop for agents
- 11:43Summary
related talks
45mBehaviour Driven Development: Embrace Uncertainty
20mBDD: You’re doing it wrong!
45mLet's talk about software behaviour (BDD)
31mBDD: Behaviour Driven Development
32mTesting in Software Development Process: TDD And BDD
15mIntro to TDD and BDD
17mUnlocking Collaboration in Code: A Journey through BDD
48mMake Your Life Simple and Carefree with BDD