AI-Guided Troubleshooting: Smarter Root Cause Analysis with OpenTelemetry and Prometheus
Modern cloud-native environments generate enormous volumes of telemetry data; traces, metrics, and logs, yet engineers still spend significant time manually correlating signals during incidents. What if AI could guide you to the root cause faster? This talk explores how to combine OpenTelemetry and Prometheus with AI-driven analysis to accelerate troubleshooting in cloud-native environments. We will walk through how OpenTelemetry’s correlated telemetry (traces, metrics, and logs) provides the structured data foundation that AI models need, and how Prometheus metrics can surface anomalies that trigger intelligent root cause suggestions. The session includes a live demo of an AI-guided troubleshooting workflow: starting from a Prometheus alert on a micro-services application, enriching the investigation with OpenTelemetry trace and metric data, and using an LLM to correlate signals and identify the root cause.
Speaker
-
Robert KingPalo Alto Networks / ChronosphereRobert is an OpenTelemetry contributor and Lead Enterprise Solutions Engineer at Chronosphere. He has spoken at AWS Summit, Cloud Native London, PyTorch Conference, and Web Performance London, most recently presenting a live demo on AI Observability with OpenTelemetry. With over a decade of experience across Datadog, Dynatrace, Chronosphere and Keysight, he has helped global enterprises operationalize cloud-native observability. Robert holds certifications in OpenTelemetry (OTCA), Prometheus (PCA), FinOps, and AWS Solutions Architect.