technology-selection

Guide .NET 8+ AI/ML technology selection across ML.NET, MEAI, MAF, and ONNX Runtime.

5.1k|377|Updated Feb 3, 2026
One-click install
npx skills add https://github.com/dotnet/skills --skill technology-selection
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: technology-selection
Source: https://github.com/dotnet/skills/tree/main/plugins/dotnet-ai/skills/technology-selection
Command: npx skills add https://github.com/dotnet/skills --skill technology-selection

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps developers select the most appropriate .NET technologies for integrating Artificial Intelligence and Machine Learning features into their applications, preventing costly mistakes and ensuring efficient implementation.

Core Features & Use Cases

  • Guided Technology Selection: Provides a decision tree to map AI/ML tasks (classification, LLM integration, RAG, etc.) to specific .NET libraries (ML.NET, Microsoft.Extensions.AI, Agent Framework, ONNX Runtime, OllamaSharp).
  • Library Layering Guidance: Explains the correct stack of libraries to use, from abstractions to provider SDKs and orchestration layers.
  • Package Recommendations: Suggests the exact NuGet packages required for each technology branch.
  • Guardrail Implementation: Outlines essential best practices for classic ML, LLM integration, agentic workflows, and RAG to ensure robustness, security, and cost-effectiveness.
  • Use Case: A developer needs to add a feature to classify customer feedback. This Skill will guide them to use ML.NET for tabular data classification instead of a more expensive and less reliable LLM.

Quick Start

Use the technology-selection skill to determine the best .NET library for classifying support tickets based on their text descriptions.

Frequently Asked Questions about technology-selection

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I choose the right .NET library for integrating LLM features like text generation and summarization?

To integrate LLM features like text generation and summarization in .NET 8+, use Microsoft.Extensions.AI (MEAI) for orchestration. This Skill guides library layering from abstractions to provider SDKs, suggesting exact NuGet packages for robust LLM integration.

What is the best way to implement a RAG pipeline with vector search in a .NET application?

The best way to implement a RAG pipeline with vector search in .NET 8+ is using Microsoft.Extensions.AI (MEAI) combined with vector search capabilities. This Skill outlines essential guardrails and package recommendations for building secure, cost-effective retrieval pipelines.

When should I use ML.NET versus an LLM for classifying customer feedback?

Use ML.NET for classifying tabular customer feedback data instead of a more expensive, less reliable LLM. This Skill provides a decision tree mapping classic ML tasks like classification, regression, and clustering to ML.NET.

Can I use ONNX Runtime for local model inference in a .NET agentic workflow?

Yes, you can use ONNX Runtime for custom local model inference in .NET 8+ applications. For agentic workflows with tool calling, this Skill guides using Microsoft Agent Framework, while ONNX handles custom model execution.

Does this .NET AI technology selection support projects targeting the .NET Framework?

No, this .NET AI technology selection requires .NET 8 or later. It does not support projects targeting the .NET Framework, pure data engineering or ETL tasks without ML/AI components, or custom deep learning training loops.

How do I run local LLM inference using OllamaSharp in a .NET project?

To run local LLM inference, this Skill guides selecting OllamaSharp as the provider SDK within your .NET 8+ application. It explains the correct library stack from MEAI abstractions down to OllamaSharp for local execution.