Reddit dev details 7 chronic failures of AI agents, from endless planning to fake progress
Nmoreland51 · reddit · 2026-11-16
A developer venting about ChatGPT, Claude and Codex on ambitious projects lists seven recurring failures: explaining instead of executing, treating solvable problems as impossible, endless planning with no progress, ignoring instructions, forcing manual coordination, losing context, and claiming completed work that doesn't actually run.
He wants Codex to coordinate Claude via computer-use tools and asks which frameworks offer persistent memory, clear responsibility splits, automatic work verification, and agent-to-agent collaboration without the human shuttling information. He's explicitly looking for real solutions, not prompts that tell the AI to "act like an engineer."
More from coding & agent
- YC F26's FlashDrive launches a shared cloud drive for AI agents — damnGruz · 2026-11-16
- Open-source coding.sh measures where your coding agents waste time and money — AsianKidTV · 2026-11-16
- Dev ships MCP server for a trading journal so Claude can query PnL and win rate — Occ_inv · 2026-11-16
- 10,000-Word Playbook: Automating Overseas KOL Marketing with AIsa and AI Agents — sujingshen · 2026-11-16
- Economist builds TeXugo, a free open-source LaTeX editor for academic papers, with Claude Code — joshgans · 2026-11-16
- Models give intelligence, harnesses give competence: coding agents cut video-editing tokens 30% — Efficient-Clock337 · 2026-11-16