Coursera Spring Sale
40% Off Coursera Plus Annual!
Grab it
Explore federated vector retrieval systems that enable secure cross-organizational data sharing for RAG (Retrieval-Augmented Generation) applications in this 43-minute conference talk. Learn how to overcome the limitations of single-organization RAG systems by implementing logical federation of vector databases distributed across multiple organizations. Discover the challenges of applying RAG across organizational boundaries, including searching for similar information across diverse sources and respecting unique data governance and access policies. Examine a practical solution that selects target databases based on metadata summaries while enforcing access control according to each provider's policies, allowing secure and efficient data utilization without physical aggregation. See a demonstration of this federated approach using a technology stack built on Milvus vector database, Open Policy Agent (OPA) for policy enforcement, and vLLM for language model integration. Understand how this methodology significantly expands the reach and value of RAG systems by enabling knowledge sharing and operational efficiency across industries while maintaining data security and governance requirements.
