Dive into the Future: .NET and AI!
summary
A runnable .NET app shows keyword search replaced by a semantic pipeline: product rows are embedded into an in‑memory vector store, queries produce embeddings and a relevance score (examples show ~82% matches and a 60% threshold), responses come from GPT-4o-mini and are postprocessed by a prompt file editable with prompty. Implementation uses Semantic Kernel for vectors and a Blazor frontend + .NET API; Net Aspire adds tracing, caching and orchestration. A separate TrafficJamAnalyzer uses GPT multimodal image prompts to classify camera frames and it demonstrably returns JSON traffic labels and timestamps.
AI-written summary · about
chapters (9)
- 0:00.NET and AI integration
- 6:03Simple .NET product search app
- 11:00Semantic search with AI
- 23:50Natural language prompts
- 31:53Fine-tuning with Prompty
- 37:53.NET Aspire scaling
- 43:35AI traffic analysis
- 50:40Debugging AI solutions
- 53:14Closing and resources
related talks
46mInfusing your .NET Apps with AI: Practical Tools and Techniques
44mSupercharge your data with Azure AI Search and OpenAI
20mBuild Intelligent Apps with .NET and Azure
50mBuilding AI ready applications
01h 10mPractical Applications of Generative AI: How to Sprinkle a Little AI in Your App
44mAI Building Blocks for .NET: Add intelligence to your C# Apps
40mUse AI for "real things" in your Windows Apps
01h 10mPractical Applications of Generative AI: How to Sprinkle a Little AI in Your App