Daniel Costea

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.