change-data-capture

Open-source data movement for ELT pipelines and AI agents — from APIs, databases & files to warehouses, lakes, and AI applications. Both self-hosted and Cloud.

Postgres, MySQL, or MariaDB to Elasticsearch/OpenSearch sync

Ecommerce Realtime Data Pipeline (Data Modeling, Workflow Orchestration, Change Data Capture, Analytical Database and Dashboarding)

Knowledge graph + MCP tool server for LLM agents with hybrid retrieval, live DB sync, Korean FTS, and memory feedback.

Repo for CDC with debezium blog post

Slowly Changing Dimension type 2 using Hive query language using exclusive join technique with ORC Hive tables, partitioned and clustered hive table performance comparison

Lightweight CDC patterns for SQLite

Sample project that describes how you can handle schema within your Django application.

Showcases real-time data replication from RDS (MariaDB) to Kinesis using AWS DMS on LocalStack. Implements both full-load and Change Data Capture (CDC) tasks to stream database changes for analytics.

Keep in sync RDB table with Hive structured store. Added Kafka as a buffer between those two tables.

Stream the data changes from Postgres to Iceberg tables

This is a tryout I prepared to demonstrate CDC (change data capture) using MySQL, Maxwell and Kafka.

Multi-cloud e-commerce data platform playground with synthetic generators, CDC pipelines, and Terraform-driven infra.

Local-first Kanban for human and AI-agent work, with Postgres CDC into ClickHouse analytics.

Change Data Capture (CDC) tool from any source(s) to any target

Transactional change feeds for SQLite

The Yelp Data Pipeline processes business reviews using Python, Kafka, AWS (DynamoDB, S3, Redshift), PySpark, AWS Lambda, and Power BI. It supports real-time streaming, CDC, daily batch processing, and data visualization for insights into customer sentiment, business performance, and industry trends.

A provider-agnostic framework to evaluate ordinary CDC (Change Data Capture) features

Setup CDC quickly.

This project create data stream from mysql using replication protocols and ingest into kafka. You can create event driven system using this.