"AI is still dumb at architecture": devs swap war stories of bad model advice
teropa · x · 2026-10-14
Developer devagrawal09 vented that he envies people working on simple enough problems to think AI is better than them at architecture — in his experience it's "just plain dumb" at architecture decisions. @0xblacklight chimed in with a favorite: Sol 5.6 xhigh tried to sell him on row-locking an entire Postgres table to "garbage collect" stale rows. The thread highlights that top models remain unreliable at high-level architectural judgment, contrasting with their everyday coding strength.
More from coding & agent
- Intent ships PR review with inline diffs, Linear integration, board views and more — LukeW · 2026-10-14
- 20 open-source web scraping and browser agent tools worth knowing in 2026 — MaryamMiradi · 2026-10-14
- JevEval uses Jev as the judge for custom LLM evals, with DeepEvent handling scoring — solyarisoftware · 2026-10-14
- Gemini 4 Argon ships in GitHub Copilot, beating prior Gemini models on every coding benchmark — rseroter · 2026-10-14
- Microsoft Copilot's real shift: from asking AI to delegating work to it — WirelessLife · 2026-10-14
- "Software is dead" debunked: vibe-coding on the fly costs tokens, compute and 15-minute waits — bendee983 · 2026-10-14