# Processing Unstructured Data with Databricks

> In this video, Eumar explains the foundation and architecture for building production-ready AI agents on Databricks across three personas: prospective students, enrolled students, and employees. He emphasizes that governance, observability, and guardrails come from Unity Catalog volumes and MLflow,…

- Web page: https://www.dataexpert.io/lesson/processingunstructureddatawithdatabricks-aug26
- Program: [Databricks AI Data Engineer Boot Camp](https://www.dataexpert.io/program/databricks-ai-data-engineer-boot-camp-2024)
- Access: Requires enrollment in Databricks AI Data Engineer Boot Camp
- Length: 51 min video
- Skills: NLP, ETL/ELT, Apache Spark, Cloud Platforms, Databricks
- Academy: DataExpert.io Academy

## About this lesson

In this video, Eumar explains the foundation and architecture for building production-ready AI agents on Databricks across three personas: prospective students, enrolled students, and employees. He emphasizes that governance, observability, and guardrails come from Unity Catalog volumes and MLflow, while a supervisor agent routes questions to specialized agents. He covers why unstructured data matters and how retrieval augmented generation works by grounding LLMs on a vector database, with chunking and embeddings handled in pipelines (often using Spark). He also discuss Databricks tools via MCP through Unity Catalog functions, and outline three build approaches: Genie for text to SQL, AgentBricks for low-code reg, and code-first custom agents, with evaluation and LLM ops planned for the next week.

## Navigation

- Previous lesson: [Setting up your Databricks App](https://www.dataexpert.io/lesson/setting-up-your-databricks-app.md)
- Next lesson: [Building Rich Context for AI Agents Day1 Lab](https://www.dataexpert.io/lesson/buildingrichcontextforaiagentsday1lab-aug26.md)