Agent engineering's last mile: models miss obvious bugs unless you tell them what to look for
MoonL88537 · x · 2026-09-18
A discussion on AI agent engineering pain points: one approach is running a static correctness pass with models like astra/fable on a narrow codepath, but you still need to know where to point it — the 'wait, what happens if these two ops run together' prompt remains on you.
The original poster agrees this is the last mile holding everything back: unless you tell agents what to look for, they won't report issues that are 'obvious' to humans; he's been stuck on this for weeks.
Related event: Hunting Race Conditions in AI-Written Code with Micro-Universe Simulations(4 posts)→
More from coding & agent
- Users beg Claude Code lead for a reset for weeks — total silence from Thibault Sottiaux — RileyRalmuto · 2026-09-18
- Vague prompts force AI to make endless calls: why good engineers still hand-write key code — bendee983 · 2026-09-18
- Codex compaction: the LLM requests a blank context and keeps its own notes — mitsuhiko · 2026-09-18
- Google's ToolGrad flips tool-use dataset generation: answers first, queries later — AxSaucedo · 2026-09-18
- Founder advice: add an MCP to your SaaS, hype or not — jonathan_wilke · 2026-09-18
- Full harness open-sourced: controller, planner, safeguards, and media verification on GitHub — imjustnewatai · 2026-09-18