Evaluating AI agents: Real-world lessons from building agentic systems at Amazon

Evaluating AI agents: Real-world lessons from building agentic systems at Amazon(aws.amazon.com)

3 points by bpedro 134 days ago | 1 comment

lumpilumpi 134 days ago |

I get the justification but I found it hard to understand how the actual evaluation at each step is carried out. For example, is there any calibration to some human gold standard involved or is the AI evaluating the AI without calibration/oversight?

lumpilumpi 134 days ago |

I get the justification but I found it hard to understand how the actual evaluation at each step is carried out. For example, is there any calibration to some human gold standard involved or is the AI evaluating the AI without calibration/oversight?