Opening Keynote: Building and Scaling Cloud Native Intelligent Apps with .NET 8
summary
Demos use .NET 8 and C# 12 features like primary constructors and collection literals, plus Azure OpenAI via the semantic kernel and the Microsoft OpenAI SDK; a chat loop that fits on-screen is a six-line chat loop, with memory + embeddings and a vector DB approach to scale retrieval. Native AOT shrinks publish sizes from ~95MB to ~9–10MB, startup to ~34ms and shows ~800k RPS-class throughput; trimmed containers reach ~8 MB containers. A local multi-project builder surfaces service URLs, logs, traces and output caching; demos mostly run with minor preview hiccups and rate limits.
AI-written summary · about
chapters (21)
- 0:00Conference opening
- 1:54Keynote speaker Scott Hunter
- 2:34.NET evolution since 2014
- 5:25C# 12 features
- 8:38Primary constructors
- 10:31Collection literals
- 12:42.NET 8 performance
- 14:51Intelligent apps with AI
- 16:43Chatbot with Semantic Kernel
- 19:09Custom AI functions
- 22:44AI memory in conversations
- 24:36Streaming AI responses
- 26:42Vector databases
- 34:58Optimizing .NET containers
- 37:29Ahead-of-time compilation in .NET 8
- 40:46AOT size reduction
- 43:17Project Tye cloud-native
- 46:24Distributed cloud-native app
- 55:11Multi-project debugging
- 59:53Azure deployment
- 1:00:15Blazer components
related talks
53mKeynote: Modern Application Development with .NET and Azure
59mBuild Next-Gen AI Apps with .NET and Azure
01h 15mPanel discussion on the future of .NET
01h 14mKeynote
44m.NET 11 in depth: Runtime, libraries, and SDK for the AI era
48mKeynote: Introducing .NET Aspire – Cloud Native Development for .NET
01h 04mOur favorite things in .NET 8
54mKeynote: Introducing .NET Aspire – Cloud Native Development for .NET