The Crucial Role of Coding Agent Harnesses
matei_zaharia · x · 2026-07-09
Databricks tested various coding agents on their internal codebase and found that harness design significantly impacts both cost and performance. A simple Pi harness achieved a success rate comparable to vendor-provided harnesses but at half the cost, partly due to a smaller context being fed into the model.
They also emphasized that cheap per-token pricing does not mean cheap per-task. Some models might be cheaper per token, but because they consume significantly more tokens, the final task cost is higher and the quality is worse.
More from coding & agent
- Alex Townsend posts 200 open problems in numerical linear algebra for humans and AI agents — IgorCarron · 2026-09-11
- Kimi K2.8 Preview rolls out: near-K3 coding performance, 1M context for all tiers — teortaxesTex · 2026-09-11
- Looking for a classifier of software engineering task shapes to pick models per task — StewartalsopIII · 2026-09-11
- Steal this idea: prompt-to-hardware where agents assemble custom devices — paraschopra · 2026-09-11
- Model Is the Least Interesting Part: A Guide to Six Core AI Architectures from RAG to Multi-Agent — goyalshaliniuk · 2026-09-11
- Non-coder builds layered memory architecture: 20k tokens tracks a year of agent conversations — matteoianni · 2026-09-11