Keynote: From Keyword Search to Vector Powered Insights: Th...
summary
OpenSearch 3.0 ships Lucene 10, native MCP and GPU acceleration and claims 9x GPU index build speed with 3x cost reduction for vector workloads; the project reports contributions from 225 organizations and added multimodal, sparse and hybrid search plus model connectors. Freshworks runs OpenSearch in production for search and observability—300–350 TB logs per day (15–20 PB retained), 14B writes and 3B reads/month—using HNSW vectors, embeddings and federated hybrid search to power working features like similar‑incidents and Freddy Agent.
AI-written summary · about
chapters (11)
- 0:00Conference opening
- 3:33OpenSearch growth milestones
- 7:32Search with vector databases
- 10:47OpenSearch 3.0 upgrades
- 15:56Cloud-native security improvements
- 18:33Freshworks’ OpenSearch journey
- 22:01Freshworks AI integration
- 31:50Hybrid search applications
- 34:38Freshworks vector search architecture
- 38:11Federated search future
- 40:17Closing remarks & Q&A
related talks
25mKeynote: OpenSearch in the Agentic Era
28mKeynote: The Next Generation of Enterprise Search: Intent‑Aware, Hybrid, and Open
28mUse OpenSearch as Your Vector Search Database To Optimize GenAI Accuracy!
06mSponsored Keynote: OpenSearch - See Everything: Open Observability for Agentic AI
44mRediscover your keyword search: Expand, Enrich and Rewrite
28mAI Agents in Action: Harnessing MCP and OpenSearch for Modern Use Cases
38mNatural Language Search-Lexical and Semantic
38mAI-Ready Data Foundation Using Amazon OpenSearch