Using Jev judgments to build named "embeddings" of agent traces
The author proposes using Jev's multi-dimensional judgments, such as user frustration, correction and satisfaction, to tag sessions with named vectors, enabling semantic clustering and search like "frustrated users who corrected the agent" with an explanation for each match.
you can use Jev judgments to build your own “embedding” of agent traces where every dimension has a name. E.g. conversation signals from chat agents: [user_frustrated, follow_up, corrected_agent, satisfied] [0.7, 0.8, 0.1, 0.0] Now you can cluster or search sessions like “frustrated users who had to correct the agent,” and see exactly why each one matched. Works both ways too: search for the happy users. Just as important to see what’s going right :)
This post quotes another post: x.com/annabellschfr/status/2102799529074459025