'My intelligence stack': developer splits AI coding work across Opus 5.5, Fable and Astra with human review gates
sh_reya · x · 2026-10-27
shreya shared her current 'intelligence stack' for building software:
- Human: creates the issue or feature idea
- Human + Opus 5.5: iterate on RFCs (the most iteration-heavy step)
- Fable tackles the RFC if it needs >4k lines of code, else Opus 5.5
- Astra reviews PRs for bugs and re-invokes the implementation agent
- Human reviews PR code and proposes simplifications (min lines added, max lines deleted), which Astra executes
- Human + Sol handle documentation
Matthew Russo quote-reacted: enterprises won't adopt software built 90+% by AI without human review—they'll just re-implement it themselves with AI. In the future, the only way to build a differentiated software product may be to write (or review) it by hand.
More from coding & agent
- Cursor launches Agent SDK: agents as code with Markdown prompts and TypeScript tools — soleio · 2026-10-27
- Vibecheck, the 'usable formal methods' hackathon, sells out two floors in SF — spikedoanz · 2026-10-27
- How do you verify AI agent outputs beyond unit tests and LLM-as-judge? — Practical_Green5775 · 2026-10-27
- AI agents may soon become the only way to run code, not just write it — IgorCarron · 2026-10-27
- Hermes Agent used as browser QA tester finds 8 issues in a bug-seeded site — techlatest_net · 2026-10-27
- Open-source FableCut lets AI agents edit a real video timeline via remote MCP — FastPresence9799 · 2026-10-27