← All IntelClip / AI AgentsWhy Claude Code won: training the model against its own harness
From Harness Engineering is not Enough: Why Software Factories Fail — Dex Horthy, HumanLayer · ≈10:13
“The difference was was that the this was the first time that a model lab trained a model against the harness that they were going to distribute it to users in.”
“if you are a uh harness builder and you don't own the model weights and you can't RL the model in your harness, you will always be at a disadvantage compared to somebody who owns both the model and the harness.”
What’s in it
- Explains why Claude Code beat earlier CLI coding agents fast
- Reveals the real edge: training models on their own harness
- Cites OpenAI's warning about harness builders without model control
Clip transcript
want to zoom out to the first great coding agent. Why did Claude Code go from nothing to 4 billion and I think now they're at 9 billion in revenue in under a year? Cuz they were great CLI agents before Claude Code. You had Aider, you had Code Buff. There was a bunch of tools in this category. They had all the same tools, read, write, edit, grep, bash. Um, so what was the difference? The difference was was that the this was the first time that a model lab trained a model against the harness that they were going to distribute it to users in. Um, and it got really, really good at this is just some of the tools, but it got really, really good at calling these sorts of tools in an agentic loop. In fact, the OpenAI team did a talk in November about basically if you are a uh harness builder and you don't own the model weights and you can't RL the model in your harness, you will always be at a disadvantage compared to somebody who owns both the model and the harness.
Comments
Sign in to comment.
Loading comments…