Post-incident review question sparks debate: did anyone actually understand the code — or is AI the one who does?
MikkoH · x · 2026-11-19
An engineer proposes adding a new question to post-incident reviews: "Did anyone know any of the code that was causing the incident?" A reply finds this somewhat scary, since the sharper alternative framing may soon be "Was the AI able to efficiently identify and mitigate the issue?" — hinting that the deepest understanding of a codebase may belong to AI rather than the team. The exchange captures growing unease about eroding human comprehension as AI writes more of the code.
More from coding & agent
- Nat Eliason: time to retire "vibe coding" — just call it coding — flowersslop · 2026-11-19
- Tencent Hy4 flash open-weights test: full data-to-deck workflow at $0.07 per 1M input tokens — JaynitMakwana · 2026-11-19
- Post-incident reviews add a new question: did anyone actually understand the code that broke? — MikkoH · 2026-11-19
- 3-Year Agent Builder Shares 5 Beginner Mistakes: Tool Bloat, Runaway Loops, Multi-Agent Traps — Fragrant-Detail-4566 · 2026-11-19
- MyClaw 3.0 launches as AI teammates with built-in tools, 5,000+ data sources and cloud compute — alifcoder · 2026-11-19
- Dirac Robotics Launches CAD-to-URDF, But a Skeptic Says It Costs a Few Euros of Tokens to DIY — kamathsblog · 2026-11-19