Understanding Evals in AI Agent Node Workflow
where do evals fit in for using the ai agent node?
is there a light eval as part of the loop in the node?
or is there a best practice for doing evals (ex. checking accuracy in knowledge base) after an ai agent node?
great write up here in your nov 2024 blog, but didn't help me understand how the ai agent node works out of the box. maybe im missing something
