Скачать или смотреть Building advanced agentic apps with PostgreSQL on Azure | BRK211

What benefits do agentic architectures bring compared to traditional RAG patterns? In this session we explore advanced agentic capabilities offered by popular GenAI frameworks (LangChain, LlamaIndex, Semantic Kernel) on top of Azure Database for PostgreSQL. Learn how to improve agentic apps by making vector search faster with DiskANN, and more accurate with Semantic Ranking, GraphRAG, and new powerful Semantic Operators. We also introduce new highly productive PostgreSQL extension for VSCode.

To learn more, please check out these resources:
https://aka.ms/build25/plan/ADAI_Devs...
https://techcommunity.microsoft.com/c...
https://aka.ms/PostgreSQLAIVideo_Buil...
https://www.wired.com/sponsored/story...


𝗦𝗽𝗲𝗮𝗸𝗲𝗿𝘀:
Abe Omorogbe
Maxim Lukiyanov


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 https://build.microsoft.com

BRK211 | English (US) | Data & Analytics


Related Sessions:
BRK204 -- https://build.microsoft.com/sessions/...
DEM564 -- https://build.microsoft.com/sessions/...
LAB360 -- https://build.microsoft.com/sessions/...


#MSBuild

Chapters:
0:00 - Introduction to the session on Building Advanced Agentic Apps with PostgreSQL on Azure
00:01:57 - Overview of agentic frameworks and their core benefits
00:16:25 - Details on Evaluation Agent's Function
00:25:22 - Improving Accuracy of Agents with Semantic Operators
00:30:24 - Implementation of Graph RAG Techniques in Azure Postgres
00:31:39 - Apache H is Ready for Production Workload
00:38:40 - Query Execution and Result Analysis
00:46:28 - Introduction to BM25 and PM 25 tables
00:46:40 - Plans for upcoming quarter - Introducing BM25 in Flex