Add Useful AI to Your Web App (Not Just Chatbots)
summary
Demo builds a smart paste button that extracts JSON to populate forms, a typeahead smart text area, and a semantic dropdown using Transformers JS + Hugging Face (BG-micro v2) for embeddings. Server-side uses Python FastAPI + transformers for zero-shot classification, Semantic Kernel for RAG against product manuals, and .NET demos with github repos (dotnet/smartcomponents, eShopSupport). Measured results show evaluator scores rising from ~0.4 to ~0.8 when manuals are available, with latency increasing from ~1.2s to ~2.1s per query.
AI-written summary · about
chapters (13)
- 0:00Intro
- 1:15Useful
- 3:581. AI at the UI layer
- 4:34Demo: Data extraction
- 9:15Demo: Typeahead
- 12:48Demo: Semantic lookup
- 19:032. In app logic
- 22:13Demo: Summarization
- 26:11Demo: Classification
- 30:20Demo: Q&A (Chatbot)
- 36:103. Eval/test
- 36:38Demo
- 41:50Outro
related talks
59mHow to add genuinely useful AI to your webapp (not just chatbots)
49mBuilding AI Enabled Web Apps
01h 10mPractical Applications of Generative AI: How to Sprinkle a Little AI in Your App
24mEmpowering Vue Apps with AI
37mBest AI Tools for Web Developers / How to Use AI in Web Development
46mInfusing your .NET Apps with AI: Practical Tools and Techniques
24mAI and Web Development: Hype or Reality
29m"AI-Powered Gamification for the Web"