Production RAG apps made easier with Astra
summary
Demoed production RAG stack uses LangChain / LLamaIndex, Amazon Bedrock, and Astra DB vector search built on Apache Cassandra and embeds hundreds of millions of Wikipedia vectors for real-time retrieval. Claims: 15x faster indexing, 12x query performance, and 18% higher relevance versus other vector stores. RAGStacks provides integration adapters and an assistant API to offload LLM work, supports multimodal content, multiple embedding strategies, and aims for low-latency, high-throughput production deployment with indexing and querying done in parallel.
AI-written summary · about
chapters (8)
- 0:00Introduction to RAG
- 2:51RAGStack components
- 8:10Astra DB for vector search
- 10:21RAGStack integration layer
- 13:12WikiChat demo
- 15:40Multimodal generative AI
- 17:06Performance benchmarks
- 19:53Workshop and resources
related talks
36mProduction Ready RAG
01h 35mTutorial: Build AI-Supercharged RAG Apps with a Vector Database
11mBuilding easily deployable RAG applications with Llamastack
53mBuild RAG from Scratch
35mThe Hidden Jewels to Solve Confabulation Challenges in Generative AI’s RAG Promise
31mSponsored Session: Building a Production Open Source RAG Application using LlamaIndex and Cassandra
47mRAG at scale: production-ready GenAI apps with Azure AI Search
30mGraph RAG: Bringing together graph and vector search to empower retrieval