After a year of agent-driven coding, the failure modes are still the same
PuzzleheadedMenu2454 · reddit · 2026-07-25
After a year of building a real product with AI agents writing most of the code, the author says the hype skips the failure modes.
The recurring problems are:
- confidently wrong code that looks fine but is subtly broken
- losing consistency across files and architecture
- failing to explain why a requested change is the wrong move
- weak handling of security, edge cases, and auth corners
- making subtle bugs worse while trying to fix them
- not knowing when to stop and back up
The takeaway is that agents are still useful, but the job shifts from typing to catching these failure modes before they ship.
More from coding & agent
- Ben Bajarin: Frontier AI is becoming the enterprise cyber-defense stack — BenBajarin · 2026-10-26
- Building AI bots on raw APIs? Devs say models are 'dumb without the harness' — madeofAjala · 2026-10-26
- HarnessMM cuts agent tokens per task by up to 88% while raising success rates — omarsar0 · 2026-10-26
- Microsoft demos intent-classifying agents built with Copilot App and Foundry, deployable across Teams and Outlook — marlene_zw · 2026-10-26
- Looking for a wearable mic with best battery life that feeds recordings to personal agents via MCP — altryne · 2026-10-26
- The identity crisis in software engineering: multiagent systems demand a new problem set — rakyll · 2026-10-26