# Building Rich Context for AI Agents Day1 Lab

> In this lab, Eumar explains how to build agent-ready foundations using Unity Catalog, Genie, and SQL-based unstructured document processing. He demonstrates Genie as a text to SQL capability that generates queries and charts, and explains how agents are exposed via MCP endpoints while access is gov…

- Web page: https://www.dataexpert.io/lesson/buildingrichcontextforaiagentsday1lab-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: 46 min video
- Skills: SQL, MLOps, LLMs, Data Modeling, Data Warehousing, Cloud Platforms, Docker, Git, APIs, Delta Lake, CI/CD
- Academy: DataExpert.io Academy

## About this lesson

In this lab, Eumar explains how to build agent-ready foundations using Unity Catalog, Genie, and SQL-based unstructured document processing. He demonstrates Genie as a text to SQL capability that generates queries and charts, and explains how agents are exposed via MCP endpoints while access is governed by Unity Catalog. He reviews Unity Catalog governance with tables, volumes, and models, including permissions, lineage, and converting Delta tables into AI search vector indexes. He creates a user-specific schema and three volumes (receipts, bootcamp docs, code) via the Databricks SDK, then generate synthetic receipts using parallel Pandas UDFs, and finally parse images and extract structured fields using AIParseDocuments and AIExtract to populate raw and parsed receipt tables.

## Navigation

- Previous lesson: [Processing Unstructured Data with Databricks](https://www.dataexpert.io/lesson/processingunstructureddatawithdatabricks-aug26.md)
- Next lesson: [Everything you need to know about Vector databases](https://www.dataexpert.io/lesson/everythingyouneedtoknowaboutvectordatabases-aug26.md)