Agentic AI Reference Architecture

Observability & Evaluation · Software component

Telemetry Change-Point Detector

Software componentObservability & EvaluationObservability & Evaluationarc:TelemetryChangePointDetector

A monitoring component that runs statistical change-point detection across all collected telemetry, including metrics no one actively tracks, to surface unanticipated behavioural changes before they compound.

Responsibility. Detects 'unknown unknown' anomalies in untracked telemetry dimensions.

Also known as: Causation-based AI detection, Unknown-unknowns detector

triggersreadsAlert Manager: triggersAlert ManagerTime-Series Metrics Store: readsTime-Series Metrics Store
Direct neighbourhood (hover for relationship types)

Relationships

reads dependency

triggers dynamic

Classification

Patterns
Statistical change-point detection
Quality attributes
Maintainability (ISO/IEC 25010)Safety (ISO/IEC 25010 | NIST AI RMF: safe)
Risks mitigated
Novel failure modes that predefined monitoring rules do not anticipate

Sources

  1. Ch10.4: T. Nguyen, "Human-in-the-Loop," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 10.4. ISBN: 9798244538229.