Dev argues most agent 'intelligence' lives in the harness, not the model
_AustinCalvert_ · x · 2026-10-21
In a widely shared take, @0xDevShah argues that much of the observed gains have come from prompting alone, and that the 'lightweight harness with minimal prompting' theory may only hold for ReAct-style coding agents. Building on his Optimus-1 swarm harness for autonomous research, he says experiments for Optimus-1.1 suggest harness engineering is barely scratched — a lot of intelligence actually lives in the harness rather than the model.
More from coding & agent
- Laude Institute launches Slingshots FOUR: 14 projects across 8 top institutions — a1zhang · 2026-10-21
- Research agent Épi sets 120 records, narrows n-queens constant interval 84x — dair_ai · 2026-10-21
- Compaction hooks in practice: keep last n messages verbatim, summarize the rest for full agent continuity — MoonL88537 · 2026-10-21
- The blank prompt box hides real operational burden: context, workflow and supervision fall on users — _AustinCalvert_ · 2026-10-21
- DeepMind paper: self-evolving agents forget as they learn — fastest learner GEPA also breaks old tasks most — rohanpaul_ai · 2026-10-21
- arXiv: Are Self-Evolving Agents Continual Learners? Plasticity comes at the cost of stability — rohanpaul_ai · 2026-10-21