How to add genuinely useful AI to your webapp (not just chatbots)
summary
Concrete patterns and demos show how to add AI into existing web UIs: smart paste fills forms from clipboard, smart combo box does semantic matching and smart text area offers role-aware completions, all implemented in Blazor/MVC/C#. Backend uses Semantic Kernel with Olama or OpenAI, Aspire for orchestration, PDFPig for ingestion, 384-dimensional embeddings and cosine search for semantic search, a RAG loop with citations, a Hugging Face zero‑shot classifier served via Python FastAPI, cached model responses for deterministic tests, and an evaluator improved from 0.61 to 0.77 (GPT‑4 reaches ~0.8) trading off latency and cost.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 10:40UI layer smart components
- 21:10Data for AI apps
- 33:50RAG for customer support
- 46:03Evaluating AI quality
- 53:22Testing and deployment
- 58:15Summary and future
related talks
42mAdd Useful AI to Your Web App (Not Just Chatbots)
01h 10mPractical Applications of Generative AI: How to Sprinkle a Little AI in Your App
46mInfusing your .NET Apps with AI: Practical Tools and Techniques
37mBest AI Tools for Web Developers / How to Use AI in Web Development
33mHow to enhance your own AI services using Semantic Kernel
49mBuilding AI Enabled Web Apps
44mSupercharge your data with Azure AI Search and OpenAI
44mAI Building Blocks for .NET: Add intelligence to your C# Apps