Scale and Orchestrate Multi-Agent Systems Effortlessly (EMEA)

Explore how to leverage multi-agent systems in your applications to optimize operations, automate recommendations, and enhance customer experience. The solution utilizes Microsoft Agent Framework, OpenAI ChatKit and NVIDIA Nemotron model on Azure AI Foundry to seamlessly connect with store databases, integrate human oversight, and deploy scalable chat agents.

This approach enables real-time analytics, predictive insights, and personalized interactions, resulting in improved decision-making, operational efficiency, and a superior user experience for application developers and users

📌 This episode is a part of a series. Learn more: https://aka.ms/AIAgentsApps/y-MSFT

Chapters:
00:08 - Welcome & Housekeeping
01:03 - Kickoff: AI Apps & Agents Dev Day Overview
02:48 - Why Agentic Workflows Matter
04:09 - What Makes an AI Agent Different from an LLM?
06:21 - Specialization and Multi-Agent Orchestration
10:03 - Introducing Microsoft Agent Framework
13:00 - Workflow Patterns: Sequential, Concurrent, Handoff & More
18:40 - Tools, Memory, and Observability in Agent Framework
21:32 - Demo: Building a Simple Workflow with DevUI
27:27 - NVIDIA AI Blueprints: What They Are & How to Use Them
33:13 - Under the Hood: Models, NIM Containers & Compute
38:12 - Financial Model Distillation Blueprint Deep Dive
42:09 - Running Blueprints on Azure vs NVIDIA Platforms
44:19 - Real-World Demo: Retail Insights & Inventory Management
53:52 - Batch Processing & Smart Restocking with AI Agents
57:08 - Q&A: Deployment, Customization & Integration
1:01:20 - Wrap-Up & Next Steps

#microsoftreactor #learnconnectbuild

[eventID:26558]

Видео Scale and Orchestrate Multi-Agent Systems Effortlessly (EMEA) канала Microsoft Reactor