What ‘harness engineering’ means for AI agents
There is active discussion around the terms “” and “” in work. The central question is whether the field is moving away from the simple idea that models should do .
The concern is whether LLMs should be allowed to decide an agent’s answers on their own, or whether the surrounding control system should matter more. The item is mainly a request to understand the term and the direction behind it, not a detailed explanation or proven method.
Key points
- “” and “” are being discussed in circles.
- The main question is whether agents are moving beyond letting models do .
- The item questions whether LLMs should directly decide every agent answer.
- For cost control, the surrounding agent structure may matter as much as the model itself.