Home

Defensive AI Security & Governance› Agent Observability›Detecting Regressions Without Ground Truth: Reference-Free Evals and Drift Signals

In this series

Agent Observability

◆OverviewStart here
1The Observability Gap: Why Traditional APM Fails on Non-Deterministic AI SystemsResearch
2The Pillars for AI: Traces, Metrics, and Evals as First-Class TelemetryResearch
3Semantic Telemetry: Capturing Reasoning, Tool Calls, and Decision Provenance with OpenTelemetry GenAIResearch
4Detecting Regressions Without Ground Truth: Reference-Free Evals and Drift SignalsResearch
5An Observability Reference Architecture: Sampling, Cost, Privacy, and the Feedback LoopResearch

HostExpert

Murali Chillakuru

Press play to begin the conversation.

0:00

10:24

Agent Observability · 4 / 5

Detecting Regressions Without Ground Truth: Reference-Free Evals and Drift Signals