Video by NetApp Instaclustr via YouTube
If you’re working on high-stakes applications like enterprise search, customer support, healthcare information, legal research, or financial services, it’s vital that an AI use provides accurate results. Hybrid RAG is a powerful approach for accuracy-critical AI use cases where factual answers matter most.
Learn from Sr. AI developer advocate David VonThenen as he builds a Hybrid RAG model using Graph + Vector RAG and Neo4j + OpenSearch.
Check the GitHub: https://bit.ly/4tTNJzv
Understand Graph RAG: https://bit.ly/4mR3ORR
Learn more about Hybrid RAG: https://bit.ly/4pz0D3b
Be sure to subscribe for all things AI!
Timestamps:
00:15: Typical vector embedding pipeline
00:35: Understanding semantic similarly
01:58: Graph-based retrieval and knowledge graphs
02:56: Hybrid RAG architecture model
03:44: Vector and hybrid-based retrieval demo with Neo4j and OpenSearch
06:34: The pros and cons of knowledge graphs and vector embeddings