Small decisions. Interesting possibilities.Submit contentSubmit

Jevals

I wanted an easier way to test my jev requests and see if my results are improving, so I built jevals! It's a local workbench for testng the Noul, Choice, and Score primitives. You can mix requests as well. You can run a seed command to see and play with some of the examples in the @typesafeai Made a quick demo video…

Added to Jevfast

How it uses Jev

Input: saved Jev requests and test cases. Jevals runs Noul, Choice, and Score requests so a developer can inspect whether results improve; it also seeds example requests. Exact Jev request and answer shape are undisclosed.

Why it is interesting

The workbench focuses on evaluating Jev requests themselves, making iteration observable without embedding a test harness in each app. The post does not show the evaluation metrics.