deliverables/retrospective.mdGreat Minds Agency — Session Retrospective
Session: April 1-3, 2026
What Worked
- Debate-then-build — 2 rounds was perfect. More debate = diminishing returns.
- Jensen on a cron — 18 reviews, 10 issues, 8 fixed. Highest-ROI agent.
- Margaret as continuous QA — 31 reports, 3 P0s caught, 7 PRs reviewed.
- PR workflow — 7 PRs through the pipeline. No broken deploys after adopting it.
- Prompt caching — one-line change, 90% cost reduction.
- SCOREBOARD.md — accountability made visible.
- Hybrid AI — Claude for brains, Cloudflare Workers AI for muscle.
What We'd Change
- QA from day one. Margaret should run from the first commit.
- PR workflow from the start. Direct pushes caused issues PRs would have caught.
- Specific nudges > generic. "Keep improving" produces cleanup nobody asked for.
- Don't let agents market unbuilt features. Honesty pass was necessary.
- System files drift fast. Dream consolidation should run from the start.
- Usage limits are structural. Haiku for sub-agents + staggered work.
- Moderator was redundant. Crons do everything Marcus did.
- Agents don't self-direct. They need task queues, not vague encouragement.
Process for Next Project
- Start with: QA + PR workflow + dream cron + SCOREBOARD + task queue
- Don't start with: moderator agent, 10-round debates, fake docs
- Always have: specific tasks in the organizer, not "find gaps"
- Never let: an agent claim a feature works without Margaret verifying
Key Quotes
- "Truth serves the work better than ambition." — Joint Summary
- "Stop building. Walk into three restaurants." — Jensen #3
- "An intelligence layer that cannot remember what worked is not actually intelligent." — Jensen #9
- "The process became the product." — Jensen #18