This repository provides a practical, hands-on course for .NET developers to learn and apply Generative AI techniques in their applications. It offers short video lessons, complete code samples, and step-by-step guidance, enabling developers to build AI-powered solutions using various models and services.

How It Works

The course focuses on practical implementation, integrating Generative AI into .NET projects using libraries like Microsoft.Extensions.AI and Semantic Kernel. It covers setup with providers such as GitHub Models, Azure OpenAI Services, and local models via Ollama, demonstrating core techniques like text generation, conversational flows, multimodal capabilities, and agent development.

Quick Start & Requirements

• Install/Run: Fork the repository to your GitHub account and utilize GitHub Codespaces for a pre-configured environment.
• Prerequisites: A GitHub account, GitHub Codespaces enabled, and a basic understanding of .NET development.
• Resources: GitHub Codespaces provides an instant, low-friction setup. Guides are available for upgrading to Azure OpenAI Services or using Ollama locally.
• Links: Generative AI for Beginners .NET, GitHub Codespaces, GitHub Models

Highlighted Details

• Five lessons covering AI basics, setup, core techniques, practical samples, and responsible AI use.
• Integrates with GitHub Models, Azure OpenAI Services, and local models via Ollama and Docker.
• Includes advanced samples for semantic search (SQL, vector, Chroma DB), real-time audio, and multi-agent applications.
• Offers multi-language support for README translations.

Maintenance & Community

Contributions are welcome via issues, pull requests, and suggestions. The repository is actively maintained by Microsoft.

Licensing & Compatibility

• License: MIT License.
• Compatibility: Permissive license suitable for commercial use and integration with closed-source projects.

Limitations & Caveats

The course primarily focuses on practical application; deeper theoretical explanations are linked externally. While local model execution is supported, it may require specific hardware configurations depending on the chosen models.