‹ BackHN Continuity

Thread

An empirical study of harness design for coding agents

225 points · 59 comments · wek

  1. vblanco · · focus · HN ↗
    This is done on Nemotron models + mistral, so its not very relevant to the current frontier of cheap chinese models + big models from Claude/GPT. Big miss not having qwen or deepseek in this research.
    1. dsiegel2275 · · focus · HN ↗
      The focus of the study was the different harness approaches and how they scale across model sizes. The fact that they used any particular set of models is irrelevant.
      1. siva7 · · focus · HN ↗
        That can only be assumed by someone with zero clue about harness engineering.

        So i can skip this study.

Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.