Engineer maps NFL offense onto agent architecture: QB as main agent, WR as tools
philipkiely · x · 2026-09-14
Engineer philipkiely proposed a playful analogy mapping an NFL offense onto AI agent architecture:
- QB = main agent, handles important work and hands off tasks
- Offensive line = the harness holding everything together
- Running back = subagent doing the hard yards cheaply
- Wide receivers = tools extending capabilities downfield
- Tight end = connectors providing support anywhere
- Playbook = skills — useful but can constrain greatness
- Head coach/OC = the user, prompting the agent
The framework resonated widely as an accessible way to explain layered agent design.
More from coding & agent
- Five open questions in consumer agents: auctions, agent incentives, and agent-hiring-agent — illscience · 2026-09-14
- Weekend hack: Project Titania reimplements Qwen3-0.6B from transformer to GPU ISA simulator — generativist · 2026-09-14
- Why AI-driven UI edits keep breaking, and the workflow one dev built to fix it — xiaosun86 · 2026-09-14
- What work are your AI agents doing without approval now? — playful_perusal · 2026-09-14
- 4-person team spends 140 minutes a day just syncing context across AI coding agents — Successful_Pilot_998 · 2026-09-14
- Hermes Agent desktop app adds separate reasoning-effort selector in composer — Teknium · 2026-09-14