​November MLflow Community Meetup

​We'll dive into:

​🔹 Claude Code & Claude Agent SDK Tracing

​Building agentic applications is complex, but it doesn't have to take weeks or months. In this session, we'll talk about how Claude Code enables rapid prototyping to test agent concepts in days, and how, with MLflow's automatic tracing for Claude Code, you get full visibility into every step your agent takes. With these traces flowing into MLflow's GenAI ecosystem, we'll show how you can seamlessly debug issues, run systematic evaluations, and iterate with confidence.

​🔹 OTEL Support for Tracing

​Gaining a unified observability view across complex, polyglot AI/ML applications has long been a challenge. When services written in Java, Go, or Rust interact with AI/ML interactions, trace data is often siloed, making it difficult to track applications end-to-end within a single platform.

​MLflow 3.6 solves this interoperability problem by doubling down on openness. The MLflow Tracking Server now offers first-class support for OpenTelemetry traces, enabling trace ingestion from any OpenTelemetry client via the OTLP protocol. This allows you to explore and search these traces in the MLflow UI or SDK using the same syntax as native MLflow traces. You can now use MLflow to track any AI/ML applications, unifying observability across your entire stack, regardless of language.

​📍 Streaming live on X, YouTube, and LinkedIn