Wiring an Agent Eval Suite Into CI
Set agent eval thresholds by observing baseline variance, not guessing.
Rosalind Wexley
The Eval Register
An Audit Trail for an Agent in a Regulated Workflow
Augustin Whitlaw
Writing Scoring Rubrics That Domain Experts Actually Agree On
Decompose criteria into checkable facts before measuring expert agreement with kappa statistics.
Malachy Winterbourne
Running a Structured Red-Team Cycle Against a Production Agent
A three-phase red-team cycle catches vulnerabilities that single-pass testing misses entirely.
Xiomara Wenzel