AI Agent and the Promotion You Didn't Ask For
summary
The talk explains how generative AI is moving from a toollike ‘copilot’ into embedded autonomous agents that join engineering teams. It covers designing a shared-context “Senior AI” (memory and depth), active tooling integration (terminals, browsers, APIs), feedback loops for continual evaluation and improvement, and the ethical/supervision limits of agent autonomy.
AI-written summary · about
chapters (15)
- 0:00Introduction to promotion
- 5:40New AI teammate
- 7:27Compounding engineering
- 9:56Harness engineering
- 14:52New role in 2026
- 17:44Stop coding directly
- 20:33Summary and results
- 21:16Q&A on agent learning
- 23:27Cognitive systems and fuzzy limits
- 26:19Early wins and failures
- 28:47Senior devs' reaction
- 29:28Model evolution
- 30:10Engineer communication
- 31:35Preparing juniors
- 34:03Context management
related talks
37mAI Agents: The New Era of Intelligent Software
40mAgentic AI: The Next Evolution of Automation
27mAgentic Engineering: Working With AI, Not Just Using It
28mAgentic AI
01h 00mKenote: AI Didn’t Replace You. It Promoted You!
20mGenAI agents - Evolution of automation
26mAI Agents as Your Co-Developer
22mAgentic AI for Engineering Leaders | Strategy & Execution Guide