Building Multi-Agents with Semantic Kernel & Azure Durable Functions | #MVPConnect
14This session introduces a cutting-edge approach to developing Durable Multi-Agents using the powerful combination of .NET, Azure Durable Functions, and the Semantic Kernel.\n\nRead More:\nhttps://aka.ms/DurableFunctionsapp\nhttps://aka.ms/CosmosDBforNoSQL\n\nThe talk will cover:\n\nHow Semantic Kernel with .NET simplifies building complex multi-agent workflows.\nPatterns like Fan-Out/Fan-In and Function Chaining for advanced workflow orchestration.\nIntegrating Azure Cosmos DB and the Semantic Kernel to create a Natural Language to SQL (NL2SQL) layer, enabling enhanced data-driven decision-making with vectorization for similarity-based queries.\nThe scalability, reliability, and security features Azure Durable Functions bring to multi-agent systems, including managed identity connectivity to Azure OpenAI and Cosmos DB.\nKey Takeaways:\nAttendees will gain insights into creating robust and scalable multi-agent systems, leveraging the latest advancements in .NET and Azure.\n\nSpeaker Bio:\nDivakar Kumar is a Microsoft MVP in AI and a Microsoft Certified Trainer (MCT), working as a Technical Architect at Flyerssoft. He actively shares his knowledge through blogs, talks, and training sessions, empowering developers to harness the potential of AI in real-world applications.\n\nSocial Handle: https://www.linkedin.com/in/divakar-kumar/\n\nPre-requisites:\nAttendees should have basic knowledge of:\n\n.NET framework and C#.\nAzure services like Durable Functions and Cosmos DB.\nConcepts of Large Language Models (LLMs) and vector similarity queries.