⇩ Markdown

considering dark software factories→considering dark software factories -- changelog→mechanisms for making agents more reliable→LLM as a judge→example of using LLM as a judge for message structure→eval models can be less powerful than application models

eval models can be less powerful than application models

evals models can be less powerful than application models

Backlinks

example of using LLM as a judge for message structure

With this specific example, the person struggled a bit with the prompt engineering because they were using a cheaper model, because generally speaking eval models can be less powerful than application models

see in context