Observability-debt-shielded AIOps for regulated finance and energy systems

In finance and energy, AIOps often fails when telemetry coverage is uneven across legacy cores, cloud services, vendors, and security boundaries. This talk introduces a practical approach to make AIOps trustworthy: telemetry-parity triage to normalize signals, quantify observability debt, and route incidents with clear confidence, and calibrated SLO risk scoring that combines service criticality, error budgets, change context, and signal quality to prioritize response. The speaker walks through an end-to-end workflow, including data ingestion, feature quality checks, triage, risk scoring, and action recommendations, shows how to reduce false positives and negatives, and outlines governance guardrails such as human-in-the-loop gates, audit trails, rollback safety, and compliance controls so automation stays safe and explainable.