Get Started with Databricks for Generative AI

The course covers details about how to deploy, operationalize, and monitor generative AI applications. The course will help you gain skills in the deployment of generative AI applications using tools like Model Serving. It also covers how to operationalize generative AI applications following best practices and recommended architectures. Finally, the course discusses the idea of monitoring generative AI applications and their components using Lakehouse Monitoring.

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

This course provides a comprehensive introduction to Lakeflow Connect as a scalable and simplified solution for ingesting data into Databricks from a variety of data sources. You will begin by exploring the different types of connectors within Lakeflow Connect (Standard and Managed), learn about various ingestion techniques, including batch, incremental batch, and streaming, and then review the key benefits of Delta tables and the Medallion architecture.

From there, you will gain practical skills to efficiently ingest data from cloud object storage using Lakeflow Connect Standard Connectors with methods such as CREATE TABLE AS (CTAS), COPY INTO, and Auto Loader, along with the benefits and considerations of each approach. You will then learn how to append metadata columns to your bronze level tables during ingestion into the Databricks data intelligence platform. This is followed by working with the rescued data column, which handles records that don’t match the schema of your bronze table, including strategies for managing this rescued data.

The course also introduces techniques for ingesting and flattening semi-structured JSON data, as well as enterprise-grade data ingestion using Lakeflow Connect Managed Connectors.

Finally, learners will explore alternative ingestion strategies, including MERGE INTO operations and leveraging the Databricks Marketplace, equipping you with foundational knowledge to support modern data engineering ingestion.

This course covers both the foundational SQL logic and advanced procedural logic in Databricks. Students will learn to leverage SQL constructs such as CTEs, temporary views, and user-defined functions, along with dynamic SQL and parameterized queries to enhance security and efficiency. The course also delves into procedural logic, where students will explore how to replicate traditional stored procedures using Lakeflow Jobs (Workflows), manage task dependencies, and automate workflows. Additionally, students will discover strategies for migrating legacy stored procedures to Databricks.