Practice
Treat agent mistakes as knowledge gaps
When an AI agent makes a mistake or gets stuck, identify what a human knows that the model doesn't, feed that context back in, and capture the correction so the agent doesn't need to ask again.
Treat agent mistakes as knowledge gaps
1 visible of 1 connections
Context
Related Concepts15
Transcript mentions
I think a useful way to think about it is anytime the model's making a mistake or doing something you wouldn't do, what do you know that it doesn't know?
Practice
Treat agent mistakes as knowledge gaps
When an AI agent makes a mistake or gets stuck, identify what a human knows that the model doesn't, feed that context back in, and capture the correction so the agent doesn't need to ask again.
Context
Related Concepts15
Transcript mentions
I think a useful way to think about it is anytime the model's making a mistake or doing something you wouldn't do, what do you know that it doesn't know?