Agentic search burns 30-50% of agent tokens: RL-trained model is 20x faster and 100x cheaper
AI Engineer · youtube · 2026-09-17
On the AI Engineer channel, Maximilian-David Rumpf of SID.ai explains why search is ripe for RL.
The problem
- Some 30-50% of an agent's tokens go to searching, almost all up front.
- Handing a task to an agent roughly doubles the odds of finding the right documents vs a search engine — at 100-1000x the cost and minutes instead of milliseconds.
Why classical pipelines can't close the gap
The rewrite→retrieve→rerank→return pipeline freezes every decision at design time with a fixed compute budget per query; the reranker often knows results don't answer the question but can't act, producing a long tail of failures patched with endless edge cases.
Why RL fits
- Verifiable reward: you either found the right document or not.
- Grindable environment: thousands of attempts per second.
- Follows the path of computer vision and chess: learned design beats human design.
Results
A specialized model runs 20x faster than a frontier model on the same task (5s vs 2min) at 1/100th the cost, keeping bad results out of the main context.
More from coding & agent
- PCH Cruise: an AI-coded racer where every Spawn world is a live git repo — majidmanzarpour · 2026-09-17
- Omarchy Linux is a dream OS for tinkering with AI agents, says mark_k — mark_k · 2026-09-17
- From babysitting simple features to full monoliths by agents in 8 months — reach_vb · 2026-09-17
- AI Engineer Search Track: 30-Year-Old BM25 Emerges as Backbone of Agentic Search — jobergum · 2026-09-17
- Innate OS Open-Sourced: Intuitive Agentic OS for General-Purpose Robots Under $1K — chris_j_paxton · 2026-09-17
- Antigravity 0-day shows how DNS rebinding on loopback turns AI coding agents into RCE — evilsocket · 2026-09-17