RAG in Microsoft Agent Framework | The Ultimate Guide to AIContextProvider

Learn how to control RAG in Microsoft Agent Framework. Master AIContextProvider and TextSearchProvider for clean, production-ready .NET pipelines. Here’s what you’ll learn in this video: 🔥 📌 Core context injection plumbing using the foundational AIContextProvider abstraction layer to cleanly hydrate model prompts. 📌 Deep-dive execution tracing across the four critical lifecycle hooks to programmatically intercept the pipeline before and after the core LLM call. 📌 Passive vs. agentic retrieval strategies using the BeforeAIInvoke and OnDemandFunctionCalling configurations inside the built-in TextSearchProvider. 📌 Multi-turn state preservation patterns linking AIContextProvider with the ChatHistoryProvider to eliminate empty StateBag data blindness. 📌 The final architectural breakdown: AIContext message injection vs. AIContext tool injection vs. standalone functions to balance single-turn latency guarantees against full pipeline lifecycle hook execution. 👉 Blog post: https://deployedinazure.com/rag-in-mi... ----- 0:00 Intro 1:15 AIContextProvider basics 7:27 AIContextProvider deep-dive 20:05 TextSearchProvider deep-dive 37:24 AIContextProvider and ChatHistoryProvider integration 46:05 AIContext message/tool vs an ordinary tool 50:42 Outro ----- Follow me 👉 🌐 Blog: https://deployedinazure.com 💻 GitHub: https://github.com/deployed-in-azure/... 🔗 LinkedIn:   / deployed-in-azure