Azure AI Search Hybrid Search: RAG Answers with Citations | Ep 7

Azure AI Search hybrid search and semantic ranker tutorial for RAG in C# and .NET 10: grounded answers with verified citations, streamed with Server-Sent Events. Episode 7 of the Kodify Dev "RAG in .NET" series. Our knowledge base is full of chunks and vectors. In this episode we use it: we search it four ways (keyword, vector, hybrid and the semantic ranker), compare the results in Scalar with the debugger, and build the part that makes RAG, RAG: a grounded agent that answers only from the documents, with verified citations, streamed token by token. Then we get the API ready for the Angular interface: conversation history, follow-up questions rewritten by a model, status events, opening the original file, and a clean Stop button. In this episode: ✅ Keyword (BM25), vector, hybrid (RRF) and the semantic ranker in Azure AI Search, and when to use each ✅ Scalar: API docs and a test client for .NET, instead of curl ✅ VectorizableTextQuery and the semantic configuration in Azure AI Search ✅ Grounding: numbered sources, a "not found" answer, and citations checked in code ✅ Guardrails in five layers: input limits, a minimum reranker score, prompt rules against prompt injection, verified citations, platform filters ✅ Server-Sent Events from ASP.NET Core: status, token, citations and done events, with a small Sse helper ✅ Follow-up questions: send the history, and rewrite the question before searching ✅ Why the debugger stops when a client cancels a request, and how to handle it in the action ⏱️ Chapters 00:00 Welcome to Kodify Dev 00:43 The plan 01:47 How retrieval works: four ways to search 03:01 Guardrails: five layers 04:06 Step 2: Add Scalar to the API 04:33 Semantic configuration in the index 05:17 Step 3: Search models 06:40 KnowledgeSearch service 09:27 Step 4: Search endpoint 10:27 Program.cs and build 11:27 Step 5: Run and compare the search modes in Scalar 13:06 Debugging the hybrid request 14:10 Semantic ranker 14:58 Step 6: Grounded agent and guardrails 16:21 Conversation history and the question rewriter 17:53 AnswerService and the minimum score 20:30 How Server-Sent Events work 22:27 The Sse helper: status and token events 23:36 Step 7-8: Ask controller and the stream 26:52 Step 9: Debug the prompt 27:24 Answer with citations 28:20 Break the guardrails: empty input and prompt injection 29:27 Streaming in Scalar 30:44 Step 10: Follow-up questions with history 31:31 Open the original file: download endpoint 32:33 Stop button: the debugger breaks on cancellation, and the fix 34:49 Production tips 35:22 Step 11: Commit and push 📚 Enterprise RAG in .NET: the full course (watch in order) Ep 1 · RAG in C# & .NET 10: live demo, tech stack and project setup:    • RAG in C# & .NET 10: Build an Enterprise R...   Ep 2 · RAG explained: embeddings, chunking, vector and hybrid search:    • RAG Explained for Beginners: How It Works ...   Ep 3 · Azure AI Foundry (Microsoft Foundry) and Azure OpenAI model deployments:    • Azure AI Foundry Tutorial: Deploy GPT Mode...   Ep 4 · Azure AI Search and Blob Storage setup:    • Azure AI Search Tutorial for RAG: Service ...   Ep 5 · Web API backend and the first AI agent:    • Microsoft Agent Framework in C#: Your Firs...   Ep 6 · Knowledge ingestion: Blob Storage, indexer, skillset and vectorizer:    • Azure AI Search Indexer in C#: RAG Ingesti...   Ep 7 · Retrieval and answers: hybrid search, semantic ranker, citations and streaming:    • Azure AI Search Hybrid Search: RAG Answers...   Ep 8 · Image RAG and image generation (coming soon) 📦 Code for this episode: https://github.com/shahidnoor19/kodif... 📝 Search service code: https://github.com/shahidnoor19/kodif... 📝 Grounded answer service code: https://github.com/shahidnoor19/kodif... 📦 Full repository: https://github.com/shahidnoor19/kodif... ▶️ Full playlist:    • RAG in C# & .NET 10: Full Course (Azure AI...   ⬅️ Previous: Ep 6 · Knowledge ingestion with the Azure AI Search indexer:    • Azure AI Search Indexer in C#: RAG Ingesti...   ➡️ Next: Ep 8 · Image RAG and image generation with agent tools 🧰 Stack: C# · .NET 10 · ASP.NET Core Web API · Azure AI Search (hybrid search, semantic ranker) · Microsoft Agent Framework · Microsoft Foundry / Azure OpenAI (gpt-4.1-mini, text-embedding-3-small) · Scalar · Server-Sent Events · VS Code + C# Dev Kit ℹ️ The keys shown in this video belong to resources that were deleted before publishing. 🔗 Connect GitHub: https://github.com/shahidnoor19 LinkedIn:   / shahid-noor-4563a6222   🔔 Subscribe to Kodify Dev for .NET, Azure, Angular and AI tutorials, and turn on notifications for the next episode. #RAG #dotnet #csharp #AzureAISearch #HybridSearch