'2,000 lines from Claude, 10-second Codex review': devs fret over AI engineering quality
_jaydeepkarale · x · 2026-11-22
Developer jaydeepkarale sarcastically notes he's stopped worrying about engineering quality at work: people generate 2,000 lines of code with Claude, have Codex review it in 10 seconds, and ask another AI agent to explain production incidents — and call it engineering. In the same thread about tiny maintainer teams, he adds that SQLite is underrated. The jab captures growing anxiety about code review depth as AI coding tools spread.
More from coding & agent
- Are Gemini 4 Neon fans actually using it in Antigravity? Coder asks — iScienceLuvr · 2026-11-22
- Cairnmind: open-source MCP server enforcing read-before-write for agent memory — Zestyclose_Bus_8031 · 2026-11-22
- Team releases Part 2 of $102,725 AI episode, sharing 50,000-character Claude skill — azed_ai · 2026-11-22
- 20GB open model builds flight simulator offline on a plane in minutes — heyshrutimishra · 2026-11-22
- Entire GTA 6 Apple Maps Site Built by GPT-6.2, Dev Says It Nails Tasks First Try — Dimillian · 2026-11-22
- Dev: Claude Failed a 3D Wall-Building Eval, Needed Manual Fixes in AI-Integrated 3D World — nptacek · 2026-11-22