Practical Applications of Generative AI: How to Sprinkle a Little AI in Your App
summary
A working demo app integrates completions API calls, OpenAI function definitions (called functions/tools), and embeddings to add AI to an existing ASP.NET Core/Blazor app with SignalR, MassTransit, EF Core, Refit and Postgres + PGVector. It keeps a 20-message window, summarizes older history to stay under token limits, implements a server-side arithmetic function to avoid hallucinated math, chains function calls (via a while loop) for multi-step tasks, and uses the Assistant beta to query uploaded PDFs with cited source locations.
AI-written summary · about
chapters (13)
- 0:00Generative AI introduction
- 1:03What is AI and generative AI
- 10:22LLM weaknesses like GPT
- 11:03Prompt engineering basics
- 17:06OpenAI Playground demo
- 20:05Building an AI chatbot
- 26:14Maintaining chat state
- 31:19GPT functions for live data
- 43:43Chaining AI functions
- 47:16Embeddings for user facts
- 54:33Domain-specific AI assistants
- 1:01:19AI applications and future
- 1:02:02Q&A
related talks
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
59mHow to add genuinely useful AI to your webapp (not just chatbots)
01h 38mIntegrating AI into Your Apps: A Developer's Guide to Generative AI APIs
41mThe future of AI and generative code Q&A
58mEasily Add GenAI to .NET Apps using Microsoft.Extensions.AI
24mHow Generative AI Apps Use APIs
24mInfuse Generative AI in your apps using