OpenSearch: Conversations with Context
summary
Shows how OpenSearch links vector indexes to LLMs (OpenAI, Anthropic, and local LLaMA3 runs) to do retrieval augmented generation with a memory index that stores conversation history. Requires a machine‑learning enabled node and Docker, registers a connector to an external model, and builds a RAG pipeline that sends selected document fields into prompts. Demo uses three months of bar stock reports vectorized for fast querying and a local "Lys" app that works offline and returns accurate document answers.
AI-written summary · about
chapters (12)
- 0:00OpenSearch intro and conversational search
- 3:10Data analysis tools evolution
- 4:35AI use barriers
- 6:40What is RAG
- 8:49RAG in OpenSearch
- 13:47RAG search pipeline
- 16:19Testing conversational search
- 19:07Beyond OpenSearch connectors
- 20:54Bar stock reports use case
- 21:59LIA local AI search app
- 24:11Future of open-source AI
- 25:16Q&A
related talks
36mBuilding Next-gen Search With OpenSearch
34mImplementing an open-source RAG with OpenSearch
28mUse OpenSearch as Your Vector Search Database To Optimize GenAI Accuracy!
47mBoosting AI With OpenSearch: Vector Stores, Similarity Search, and Retrieval-Augme...
39mExperiments with OpenSearch and AI
32mOpenSearch Agentic Memory: Intelligent Context for AI Agents
30mBeyond Keywords: Building an AI-Powered Search That Reads Your Mind
21mBuilding Privacy-First Search With Containerized OpenSearch and Local LLMs