On .NET Live | Building agentic systems with local-first SLMs
Featuring: Daniel Costea, Cam Soper, Frank Boucher, Maria Wenzel #dotnet
Agentic AI is an essential paradigm for enterprise automation, enabling developers to create agents that reason, act, and collaborate in .NET workflows with Microsoft Semantic Kernel. This hands-on session shows how to build agentic systems with local-first SLMs, either via Ollama server using OllamaSharp or fully in-process with LLamaSharp. We’ll demonstrate Retrieval-Augmented Generation (in-memory vector store + embeddings) for grounded answers, output evaluation with Microsoft.Extensions.AI.Evaluation, and Semantic Kernel filters for deep observability and auditing. You’ll see how SLMs deliver lower latency, reduced action costs, stronger data control, and ease of deployment for edge or on-prem environments. Architecture is exemplified with live demos of real-world processes, highlighting tool use, state management, and safe recovery loops. Scenarios include autonomous execution and human-in-the-loop approvals/escalations. Attendees will leave with concrete deployment patterns, working C# examples, and practical guidance for making SLM-first agentic AI cost-effective and production-ready.
Featuring: Daniel Costea, Cam Soper, Frank Boucher, Maria Wenzel #dotnet
Agentic AI is an essential paradigm for enterprise automation, enabling developers to create agents that reason, act, and collaborate in .NET workflows with Microsoft Semantic Kernel. This hands-on session shows how to build agentic systems with local-first SLMs, either via Ollama server using OllamaSharp or fully in-process with LLamaSharp. We’ll demonstrate Retrieval-Augmented Generation (in-memory vector store + embeddings) for grounded answers, output evaluation with Microsoft.Extensions.AI.Evaluation, and Semantic Kernel filters for deep observability and auditing. You’ll see how SLMs deliver lower latency, reduced action costs, stronger data control, and ease of deployment for edge or on-prem environments. Architecture is exemplified with live demos of real-world processes, highlighting tool use, state management, and safe recovery loops. Scenarios include autonomous execution and human-in-the-loop approvals/escalations. Attendees will leave with concrete deployment patterns, working C# examples, and practical guidance for making SLM-first agentic AI cost-effective and production-ready.
Show Playlists
