Architect MD workflow: Gemini one-shot outputs that beat Opus, via 5-lens plan critique
oikk01 · reddit · 2026-10-16
A Redditor shares a Gemini workflow producing remarkable one-shot outputs (demo links: an interactive F1 car dissector, a single-prompt Rocket League game), claiming results clearly better than no-workflow runs and even beating Opus 5.5 on the same prompt — at heavy token cost.
The core is an "architect MD" instruction file: the agent acts as head architect, first writing down the task, deliverables, what an expert would produce, and expected difficulties, then drafting a detailed architecture plan and iterating on it (a taste agent is being experimented with) before locking it. The plan must be critiqued through five lenses:
- Goal achievement — end-to-end, no missing implicit prerequisites.
- Expert-level output with concrete executable verification checks (CLI checks, headless frame-stepping, numerical assertions, FPS budgets) run by an evaluation subagent.
- Tasteful implementation & restraint — audit for visual/UI/architectural bloat.
- Anticipated difficulties with exact mitigation plans written into the spec.
- Edge cases with code guardrails and tests for each.
After locking, subagents implement, then multiple test subagents validate and the app iterates until scores are high.
More from coding & agent
- Why company AI agents go stale: adoption needs in-house knowledge, not just better agents — EdenEmarco177 · 2026-10-16
- Cheng Lou open-sources Handwired: hand-written weights turn code into training-free neural nets — dotey · 2026-10-16
- Building a portable context vault in Notion so every AI agent knows you from day one — ivanhzhao · 2026-10-16
- Lovable apps can now talk to customers on WhatsApp with inbox, reminders and templates — AlexandrePesant · 2026-10-16
- FDE-Gym benchmark: best agent fully solves only 14 of 95 real customer-delivery tasks — dair_ai · 2026-10-16
- Deep Research API showdown: OpenAI, Perplexity, Parallel, Tavily, Gemini, Valyu compared — bar_raiser333 · 2026-10-16