Developer Postmortem: Fully Autonomous AI Coding for Complex Projects Fails
MarcJSchmidt · x · 2026-07-25
A developer shared lessons learned from six months of using AI tools like Claude, GPT, and Kimi for fully autonomous coding, which ultimately failed. For non-standard, complex projects, letting AI take the wheel leads to a decayed codebase filled with inconsistent files and drifted assumptions.
He has since shifted to a human-in-command strategy: he now writes the core code and pseudo-code himself, using AI only to fill in the blanks. He also intentionally writes documentation in his own language—being purposefully vague or specific—to maintain 100% understanding of the project. He also noted the struggle of finding the perfect claude.md prompt to stop the AI from overstepping and wrecking the codebase.
More from coding & agent
- A Company Says It Likes Coding Agents but Won’t Put Them in the Core Product — remilouf · 2026-07-25
- Why most enterprises should build on agent platform components, not the platform itself — bibryam · 2026-07-25
- What real long-running agents are actually useful in business workflows? — lenn_rt · 2026-07-25
- Go project mcpconsole adds an embeddable web UI for MCP servers — bilus · 2026-07-25
- Anthropic says its agent platform is built for self-hosting, MCP tunnels, and external runtimes — dsp_ · 2026-07-25
- How to handle captchas when running Claude Code and Codex automation — dry_Relationship2007 · 2026-07-25