Developing AI Agents with DSPy and MLflow in Databricks

We continue exploring the capabilities of [DSPy](https://dspy.ai/) for building LLM-based applications\*. We add monitoring from [MLflow](https://mlflow.org/genai/observability) and simplifications from [Databricks](https://docs.databricks.com/aws/en/machine-learning/). (\* For now, I’ll refrain from using the term "agent" outside the title.) What you’ll learn (you’ll discover while coding live): 1. You’ll create a **DSPy** project with **MLflow** (using the good old [uv](https://docs.astral.sh/uv/) already at version [0.9.16](https://github.com/astral-sh/uv/releases/tag/0.9.16)! 🔥) 2. You’ll get familiar with the **MLflow CLI** 3. You’ll explore [MLflow tracking](https://mlflow.org/docs/latest/ml/tracking/) for monitoring AI applications (OSS MLflow and Databricks-managed MLflow) 4. You’ll build an autonomous AI application with **DSPy** (**dspy.Signature**, **dspy.Module**, and the "highlight moment" **dspy.ReAct**) If you have questions, suggestions, or would like to help, please contact me at jacek@japila.pl. Any assistance is greatly appreciated! Please spread the word about this meetup among your friends. Thank you and see you there!