Why This Course?

Our highly rated MLOps course has helped hundreds of practitioners learn how to take ML models from prototype to production with confidence. Many of these principles apply to LLM applications.

LLMOps is not just about prompt engineering. It’s about applying proven MLOps principles to the new challenges of LLM applications:

- Reproducibility and traceability of LLM applications.

- Reliable deployment & monitoring.

- Managing cost, performance, and safety in production.

What You’ll Learn

We’ll teach you how to bring LLMs into production on Databricks while following modern software engineering and LLMOps best practices.

In the course, we will:

✅ Use MLflow Tracing and Evaluation to evaluate and trace LLM applications.

✅ Log agents using MLflow and register them in Unity catalog.

✅ Use MLflow Prompt optimization and Prompt Registry.

✅ Utilize Databricks Vector Search Index.

✅ Deploy agents using Mosaic AI Model Serving

✅ Deep dive into managed Databricks MCP, and custom applications

✅ Define and deploy resources using Databricks Asset Bundles

✅ Apply CI/CD and DevOps patterns to LLM development.

This course builds directly on the lessons from our MLOps course but is tailored for the unique challenges of LLM applications.