Agents should fit their harnesses to data, not hand-code the workflow layer
BraceSproul · x · 2026-07-23
The post argues that agent harnesses should be learned from data, not hand-designed.
- Just as neural network weights are discovered by fitting to data, the harness around an agent should be optimized from real workflow data.
- As more intelligence moves in-house, the harness becomes more valuable because it encodes business-specific details.
- The suggested process is careful data curation, auto-research, a holdout set, and human review.
- The underlying claim is that this makes agents more token-efficient and generally better.
More from coding & agent
- Cognition's SWE-2 uses a KKT duality argument in RL to shift the effort Pareto curve — YouJiacheng · 2026-09-11
- First-ever Three.js Conference lands in Paris, with a panel on AI-shortened design workflows — OdinLovis · 2026-09-11
- Data engineering, not agent frameworks, is the real bottleneck for enterprise AI agents — dhruv2038 · 2026-09-11
- RTK Terminal Compression Cuts Tokens but Leaves Your AI Coding Bill Unchanged — Bartaseth · 2026-09-11
- GPT-6 Astra beats Factorio with enemies in 44 in-game hours at ~$4,500 API cost — liminal_bardo · 2026-09-11
- Investment Analyst Asks How to Build a Claude-Based Diligence Agent Stack — Careless_Tie2286 · 2026-09-11