Humans on the loop, not in it: Taking agentic engineering end to end
summary
Argues for an all‑in agentic workflow where agents (e.g. Claude) write and iterate code without humans editing source, and continuous delivery becomes a pipeline‑as‑operability‑harness that deploys, runs integration tests, and measures outcomes. Experiments show agents frequently fail on infrastructure deployment, loop on expensive retries, and patch environments instead of fixing source, so full automation is brittle. Practical controls include scripted agent skills, linting/tests, mutation testing and observability tied to DORA metrics to enforce durable, operable software.
AI-written summary · about
chapters (9)
- 0:00Humans on the loop intro
- 3:24Touching code defeats agents
- 9:23Humans in loop to pipelines
- 14:16Good software for agents
- 19:31Measuring deliverability and operability
- 23:45Agents vs DevOps delivery
- 26:33Pipelines as operability harnesses
- 30:48Agent deployment challenges
- 31:30Summary and future
related talks
33mHumans on the loop, not in it: Taking agentic engineering end to end
27mAgentic Engineering: Working With AI, Not Just Using It
34mThe Human-in-the-Loop is Tired
35mThe Agentic AI Engineer
46mAgentic Engineering: From Writing Code to Reviewing It
21mPlatforms for Humans and Machines: Engineering for the Age of Agents
46mWhen You Run Out of Requirements: What Happens When You Go All-In on Agents
30mBuilding Tools with Tools: Modern AI-Driven Development