Beyond the code: Leading engineering teams in the GenAI era
summary
Argues AI takes teams about 70% Problem of the way but the last 30% — edge cases, architecture fixes, tests — still needs human expertise; cites Google data: nearly a third of accepted code originates from AI, developers complete ~21% more tasks and merge ~98% more PRs while PR sizes grow +154% and review times +91%. Recommends context engineering, spec-driven development, a learnings markdown memory file, tools like Gemini CLI and Chrome DevTools MCP, and pilots that measure cycle time, quality, and developer trust.
AI-written summary · about
chapters (15)
- 0:00AI's impact on engineering
- 2:17AI as standard tool
- 5:49Piloting AI tools
- 8:40Productivity gains from AI
- 10:03Context engineering for AI
- 11:48Spec-driven development
- 13:34Prototyping with AI
- 15:38Tests and docs before AI
- 17:03AI adoption reality check
- 19:55AI coding strengths and limits
- 20:17Preventing skill erosion
- 23:24Fixing AI measurement problems
- 24:47AI governance and guardrails
- 25:29Human skills in AI era
- 25:51Change management for AI
related talks
38mFrom code to culture: How AI is reshaping the engineering organization
27mEngineering in the Age of AI: Making AI Work for Us, Not Replace Us
21mAI Code Quality Debt: What Strong Engineering Teams Do Differently
36mEngineering Intent: Solving the AI Productivity Paradox
18mHumans + AI: The Future of Software Development Teams
47mAI, Automation & DevEx: Fueling High-Velocity Engineering
25mThe AI-Native Software Engineer
01h 01mAre Your Engineering Teams Prepared for the GenAI Revolution