Using a large model as tech lead: speculative-decoding-style agent orchestration
prajdabre · x · 2026-09-26
The author shares his favorite agent pattern: a larger model orchestrates, dispatches sub-agents for subtasks, verifies their output, and ties up loose ends. He likens it to speculative decoding — a cheap sub-agent "drafts" an action and the expensive main agent verifies it.
In his experience the main model is a good "tech lead": if sub-agents mess up, it says "fine, I'll do it myself." Letting one big model do everything — task, verification, goal-checking — goes wrong more often. He also mentions an acquaintance simulating a 3-tier org where they act as VP with director and tech-lead layers beneath.
More from coding & agent
- From Automation to Automated Automation: A 40-Minute Dive Into Agentic AI's Real Risks — DavidLinthicum · 2026-09-26
- Developer Suspects Cost-Driven Serving Optimizations Are Breaking Multi-Step LLM Reliability — karmay007 · 2026-09-26
- mobile-mcp: MCP Server for Mobile AI Automation Hits 7K+ Stars on GitHub — mobile-next · 2026-09-26
- Redis City: A 3D Visual Explainer That Walks You Through Redis Internals — JeremyCMorgan · 2026-09-26
- The four types of vibe coder: Observer, Muser, Architect, Savant — GrungeWerX · 2026-09-26
- The automation ran fine but did nothing: put the target on the workflow, not the run — sam03069 · 2026-09-26