Engineering in the Age of AI: Making AI Work for Us, Not Replace Us
summary
Claims concrete: an AI pair-programmer (Claude, paid AI Max) cuts coding time by about 50% faster development, but output multiplies pull requests and bugs, overwhelming CI/CD and requiring new automation. Teams clean docs, standardize domain language, adopt outbox/event-driven and DDD choices, and double down on mentoring to recover lost learning. Running five agents in parallel (example: Guest Town orchestration) speeds creativity but removes recovery time and creates mental burnout unless workstreams and review gates are limited.
AI-written summary · about
chapters (10)
- 0:00AI and engineering fears
- 4:45AI as coding assistant
- 8:21AI adoption challenges
- 11:59Mentoring and social learning
- 13:22Cleaning knowledge for AI
- 15:31Scaling with AI output
- 17:16Shift to strategic discussions
- 19:24Human judgment over quantity
- 21:14Burnout from AI workload
- 23:03Key takeaways and future
related talks
18mHumans + AI: The Future of Software Development Teams
27mAgentic Engineering: Working With AI, Not Just Using It
26mBeyond the code: Leading engineering teams in the GenAI era
01h 00mKenote: AI Didn’t Replace You. It Promoted You!
01mAI Is Changing How We Developer Software!
32mAI and Engineering Closing the Gap Between Expectations and Reality 1
30mEngineering Fundamentals in the Age of AI: Why Human-in-the-Loop Matters More Than Ever
10mInvesting In AI Era