ML.NET

Microsoft's open-source machine learning framework for .NET. Unlike Semantic Kernel (which focuses on LLM integration), ML.NET is for traditional machine learning: classification, regression, anomaly detection, recommendation, text analytics. It includes AutoML capabilities for automated model selection and hyperparameter tuning, plus a Model Builder tool in Visual Studio for code generation.

Why it matters for writers: ML.NET predates the LLM era and solves a different class of problems. If your AI project needs traditional ML capabilities (sentiment analysis, spam detection, tabular data prediction) alongside or instead of LLM features, ML.NET is the .NET-native option. The important distinction: ML.NET and Semantic Kernel are complementary, not competing. ML.NET classifies and predicts. Semantic Kernel coordinates LLMs. They can coexist in the same application like two roommates who have different jobs and share a kitchen.

Related terms: Semantic Kernel · ONNX Runtime · NuGet