James S.

Phoenix is very flexible, it took me some time to figure out a workflow. Anyway, I just finished the first round evaluation on our AI feature, and generated a CSV report, hoping the explanations in the annotations are helpful to our AI dev team.

Roger Y. Wow, cool! I’ll use your method when I run experiments.

Just finished the script: https://gist.github.com/goofansu/5402aab11573262e3da02c718806b4ad

Thank you!

Roger Y. Thank you. I need feedback because I want to export the root spans with explanations to a CSV for the other team to improve the AI feature.

Roger Y. I changed to evaluate the AGENT spans instead of run experiment for my requirement.

Now there are feedbacks with explanations in each span. Now my understanding is:

Correct me if I’m wrong. Thanks

Dustin N. No worry, maybe I just use the wrong way to log feedback. Roger Y. I also think it’s not necessary to log_evaluations manually in the evaluator. But if I just run evaluation in experiments and without log_evaluations, there is no feedback in the evaluator traces. For feedback I mean the screenshot:

Yes, the most interesting is as a Rails developer, I found Phoenix is very intuitive to learn and enjoyable to use. Thank your team for the great work!