Tips for fast vector and full-text search with Azure Cosmos DB | DEM567

Enable smarter and more intuitive information discovery in AI apps and analytics with Azure Cosmos DB. This coding session will show ways to balance accuracy and speed with full-text search for precise keyword matching; vector search for semantic understanding that captures the meaning behind queries, and hybrid search which uses the strengths of both to deliver enhanced relevance and contextual accuracy.

To learn more, please check out these resources:
aka.ms/CosmosDB/VectorSearch
aka.ms/CosmosDB/BillionScaleVectorSearch


𝗦𝗽𝗲𝗮𝗸𝗲𝗿𝘀:
James Codella
Haiyang Xu


𝗦𝗲𝘀𝘀𝗶𝗼𝗻 𝗜𝗻𝗳𝗼𝗿𝗺𝗮𝘁𝗶𝗼𝗻:
This is one of many sessions from the Microsoft Build 2025 event. View even more sessions on-demand and learn about Microsoft Build at build.microsoft.com/

DEM567 | English (US) | Data & Analytics


#MSBuild

Chapters:
0:00 - Features and Capabilities of Azure Cosmos DB
00:01:34 - Tips for Efficient Data Ingestion with Cosmos DB
00:07:07 - Details on Full Text Contains Search Operation
00:07:24 - Explanation of Full Text Scoring Using BM25 Method
00:07:43 - Executing Query Syntax and Introducing Hybrid Search
00:08:47 - Discussion on Billion Scale Vector Search Capabilities
00:09:42 - Enabling Better Throughput with Client-Side Execution Options
00:12:34 - Recap and Code Example Demonstration
00:13:12 - Transition to F