Stop building agents as a straight line: the 90-second test to cut fake dependencies
blaizedsouza · x · 2026-09-27
Reposting @imryven's argument: most agents are still orchestrated as a straight line ("do A, then B, then C"), even though half the steps don't actually depend on the previous one — so you pay for the wait on every run. He proposes a 90-second check before writing any orchestration code: walk the workflow step by step and ask at each seam whether the step truly needs the last one's output; keep real edges, cut fake waits.
Split what remains into a scaling pattern:
- Fan out: one agent per independent piece, all running at once
- Reduce: plain code, zero model tokens — just flatten and dedupe
- Verify: a fresh agent with no memory of the work, one job: try to kill it
- Synthesize: a final agent writes the answer
The quoted tweet links to his article "Graph Engineering: My System Checked Its Own Work. It Was Lying About It.", born from frustration with babysitting an agent that forgot what it was doing every twenty minutes.
More from coding & agent
- FuSheng: AI now navigates human interfaces like a person and maintains real relationships — FuSheng_0306 · 2026-09-27
- AI agent sends personalized Mid-Autumn greetings to 50 contacts by seeing the screen and clicking, then replies to 70+ messages — FuSheng_0306 · 2026-09-27
- FuSheng had GPT-6 take over his WeChat and send 120+ error-free holiday greetings — FuSheng_0306 · 2026-09-27
- FuSheng let GPT-6 run his WeChat and send 120+ flawless holiday greetings — FuSheng_0306 · 2026-09-27
- Pairing Astra and Opus for coding: 3-stage pipeline tops at 82.8, but solo runs cost far less — kevinkern · 2026-09-27
- UT Arlington paper: AI agents miss requirements even when marking tasks complete — alex_verem · 2026-09-27