Skip to Content
View all events

Change Data Capture in Action with SQL Server & PostgreSQL

Published by O'Reilly Media, Inc.

Intermediate content levelIntermediate

A practical guide for building a real-time data layer for analytics, applications, and AI

What you’ll learn and how you can apply it

  • Design systems that capture database changes and route them reliably across modern data systems
  • Develop strategies for different CDC implementation approaches and patterns that fit specific use cases
  • Integrate CDC change streams into AI systems to keep RAG contexts and agent memory fresh

Course description

Industry surveys show that nearly three-quarters of organizations rely on streaming or real-time data in critical systems. As operational data becomes more important to analytics, services, and AI, teams need reliable ways to capture and route changes without adding unnecessary complexity to the source system.

This course explores Change Data Capture, or CDC as an architectural decision for building low latency pipelines powering near real-time systems. By using an operational database as a change source, teams can deliver fresher data to downstream consumers while avoiding unnecessary work on source systems. You’ll learn how CDC supports analytics, applications, and AI workflows that enable organizations to make informed decisions faster, and how to evaluate the trade-offs among different approaches.

This live event is for you because...

  • You’re a data, platform or analytics engineer or architect who builds or manages pipelines and operational databases.
  • You are a key contributor to the real-time data strategy for your organization.

Prerequisites

  • Basic to Intermediate knowledge of SQL
  • Basic exposure to streaming concepts like event-driven architecture
  • Familiarity with relational databases like PostgreSQL and Microsoft SQL Server

Recommended follow-up:

Schedule

The time frames are only estimates and may vary according to how the class is progressing.

Understanding Change Data Capture (20 minutes)

  • Presentation: Building a CDC decision framework
  • Discussion: Name one thing you work with that pulls from a database on a schedule. Describe why that interval was good enough for your team; why did you not need real time?
  • Exercise: Match business scenarios with a CDC approach

CDC with SQL Server (35 minutes)

  • Presentation: Enabling and reading SQL Server CDC
  • Exercise: Enable CDC and read the change table
  • Break
  • Q&A

CDC with PostgreSQL (35 minutes)

  • Presentation: Logical decoding, replication slots, and the Debezium event envelope
  • Exercise: Start Debezium Server and inspect the event envelope
  • Break
  • Q&A

Routing Change Data to Consumers (25 minutes)

  • Presentation: CDC for AI: Data freshness for RAG and agent memory
  • Exercise: Route a change event to consumers using an AI assistant
  • Discussion: What's the first thing in your stack you'd let an agent read live and what would make you nervous about giving it that access?
  • Q&A

Connecting Agents to Change Data with MCP (25 minutes)

  • Presentation: MCP and connecting agents to change data
  • Exercise: Trace a change event into agent memory
  • Break
  • Q&A

Additional Topics and Considerations (20 minutes)

  • Presentation: CDC in production: Considerations and Operations
  • Discussion: Tell us about the last time something broke in a pipeline you owned, what advice would you give someone about what you learned from your experience?
  • Q&A
  • Closing

Your Instructor

Jasmine Greenaway

Jasmine Greenaway is a developer relations leader with 15+ years spanning software development, technical advocacy, and education. Her current work focuses on databases in the modern development stack and the experience of developers who are building with them. She specializes in experimental content and engagement strategies that reach new audiences and helps make technical content more accessible.

She’s currently a developer advocacy manager at Datadog and an adjunct lecturer in computer science at City University of New York. She’s also author of Fundamentals for Self-Taught Programmers (Packt Publishing), helping aspiring developers bridge the gap from beginner to practitioner. She holds an MS in software engineering from University of West Florida and a BS in software engineering from Embry-Riddle Aeronautical University.

Skills covered

  • SQL
  • Data Lake
  • Exploratory Data Analysis
  • Data Center