Retroalimentation among agents / Iterative agent system

Crews
, , ,
1

Hello! :waving_hand:

I’m new to CrewAI, and I’d like to create a workflow where agents can provide feedback to each other in a loop.

For example, if I have a Writer_Agent and a Feedback_Agent, I want to set up a system where the Writer_Agent improves the work based on the feedback provided by the Feedback_Agent. This loop would continue until the Feedback_Agent determines that the output is no longer improvable.

Is this possible to achieve in CrewAI? If so, what would be the best way to implement it?

Thanks in advance! :blush:

2

I suggest you take a good look at this video here by @bhancockio. It covers Flows with Feedback Loops.

1 Like
3

Thank you for your response! :blush: I’ll check it out!

1 Like