AI-Ready Data Foundation Using Amazon OpenSearch
summary
Shows a working OpenSearch pipeline that ingests PDFs, emails and logs, creates vector-enabled indexes, and serves hybrid semantic plus keyword queries with subsecond latency and built‑in reranking and relevance tuning. Demonstrates RAG architectures integrated with Amazon Bedrock and SageMaker for embedding and model hosting, conversational memory for chat context, and agent-style multihop reasoning that autonomously refines queries. Demo on Hugging Face retrieves and filters shoe catalog entries (semantic intent then keyword filter) and returns correct results in real time.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 1:47AI-ready data foundation
- 5:19Why AI-ready foundations matter
- 8:08Challenges in AI foundations
- 10:35Amazon OpenSearch for AI
- 13:22OpenSearch for AI apps
- 17:13Search evolution in AI
- 18:41Sparse vs dense retrieval
- 20:45Conversational search & RAG
- 22:54AI agents in data processing
- 29:53Production conversational search
- 32:21Hybrid search & agent demo
- 37:20Q&A
related talks
43mHarnessing DeepSeek-R1 and OpenSearch for RAG-Powered Intelligent Info...
27mElevating AI Applications with OpenSearch's Flow Framework and RAG Tool
30mBeyond Keywords: Building an AI-Powered Search That Reads Your Mind
47mBoosting AI With OpenSearch: Vector Stores, Similarity Search, and Retrieval-Augme...
38mBuild AI-Ready Search: Integrating Docling with OpenSearch for Advanced RA...
28mImproving responses in generative AI applications
45mMaking semantic search & RAG real: How to build a prod-ready app
26mOpenSearch: Conversations with Context