Building Retrieval Agents On Databricks

This course provides a comprehensive introduction to Lakeflow Connect, a scalable and simplified solution for ingesting data into Databricks from a wide range of sources. You’ll begin by exploring the different types of Lakeflow Connect connectors (Standard and Managed) and learn various data ingestion techniques, including batch, incremental batch, and streaming ingestion. You'll also review the key benefits of using Delta table and the Medallion architecture

Next, you’ll develop practical skills for ingesting data from cloud object storage using Lakeflow Connect Standard Connectors. This includes working with methods such as CREATE TABLE AS SELECT (CTAS), COPY INTO, and Auto Loader, with an emphasis on the benefits and considerations of each approach. You’ll also learn how to append metadata columns to your bronze-level tables during ingestion into the Databricks Data Intelligence Platform. The course then covers how to handle records that don’t match your table schema using the rescued data column, along with strategies for managing and analyzing this data. You’ll also explore techniques for ingesting and flattening semi-structured JSON data.

Following this, you’ll explore how to perform enterprise-grade data ingestion using Lakeflow Connect Managed Connectors to bring in data from databases and Software-as-a-Service (SaaS) applications. The course also introduces Partner Connect as an option for integrating partner tools into your ingestion workloads.

Finally, the course wraps up with alternative ingestion strategies, including MERGE INTO operations and leveraging the Databricks Marketplace, equipping you with a strong foundation to support modern data engineering use cases.

Note: For SCORM lecture files, please ensure that you close the SCORM window after completing the content. Do not click the ‘Next Lesson’ button, as doing so may prevent the SCORM module from being marked as complete.

After consuming the content inside of this learning pathway, you should be able to deploy, operationalize, and monitor generative deploying, operationalizing, and monitoring generative AI applications. This content will help you gain skills in the deployment of generative AI applications using tools like Model Serving. We’ll also cover how to operationalize generative AI applications following best practices and recommended architectures. Finally, we’ll discuss the idea of monitoring generative AI applications and their components using Lakehouse Monitoring.

Note: 

1. This is the fourth course in the 'Generative AI Engineering with Databricks’ series.

2. Databricks Academy is transitioning from video lectures to a more streamlined PDF format with slides and notes for all self-paced courses. Please note that demo videos will still be available in their original format. We would love to hear your thoughts on this change, so please share your feedback through the course survey at the end. Thank you for being a part of our learning community!

This course introduces Databricks Genie, a conversational analytics tool that lets you ask plain-language questions of your organization's data and get governed, secure answers. You'll learn what a Genie Space is, how it keeps your data secure, and best practices for getting the best results. The course closes with a hands-on game where you will compete to investigate a business mystery.