Retrieval-Augmented Generation (RAG) is transforming analytics applications — but implementing it often means managing multiple systems: OLTP, vector DBs, and orchestration tools.
In this session, we’ll show how OceanBase simplifies this stack by supporting both structured and vector data natively, enabling developers to build real-time RAG pipelines using just one open-source database.
We’ll walk through a working demo that combines OceanBase with OpenAI and popular Python frameworks like LangChain, demonstrating how to perform vector search and retrieval directly using SQL.
Unlike traditional setups that require combining a relational database and a separate vector database, OceanBase handles both transactional and semantic search in a single engine — with consistency, availability, and simplicity.