Quiz 12 Question 8 of 20

A multi-agent claims team needs continuous quality scoring on thousands of nightly agent runs. Human review can cover only a small sample. They want automated semantic scoring of final answers against policies, with periodic human calibration, feeding a semantic optimization loop. Which approach best fits Foundry-oriented continuous improvement?

Select an answer to reveal the explanation.

Motivation