Recall the GQM moves that keep agent evaluation tied to decisions.
Core memory GQM What does GQM force you to define? Recall the chain before metrics. Reflex vs move Metric-first or goal-first? GQM does not reject metrics; it makes them accountable to a question. Question test What makes a GQM question useful for an agent eval? It asks about observable behavior that changes the decision: tool use, policy boundary, clarification, escalation, evidence quality, or recovery. A weak question asks whether the agent is good. A strong question asks whether a behavior holds in the target workflow. Dashboard objection Can we just track all the metrics and decide later? The team wants…
Sign up free — one personalized lesson every day, matched to your role and goals.
Already have an account? Sign in