GenAI for .NET: Build LLM Apps with OpenAI and Ollama

In this hands-on course, you'll learn to integrate OpenAI, Ollama and .NET's new Microsoft-Extensions-AI (MEAI) abstraction libraries to build a wide range of GenAI applications—from chatbots and semantic search to Retrieval-Augmented Generation (RAG) and image analysis.

Throughout the course, you’ll learn:

.NET + AI Ecosystem

You'll learn about Microsoft's new abstraction libraries like Microsoft-Extensions-AI, which makes it super easy to integrate & switch different LLM providers like OpenAI, Azure AI, Ollama and even self-hosted models.

Setting Up LLM Providers

Configure the LLM providers—such as GitHub Models, Ollama, and Azure AI Foundry—so you can choose the best fit for your use case.

Text Completion LLM w/ GitHub Models OpenAI gpt-5-mini and Ollama llama3.2 Model model

You’ll learn how to use .NET to integrate LLM models and performs Classification, Summarization, Data extraction, Anomaly detection, Translation and Sentiment Analysis use cases.

Build AI Chat App with .NET and gpt-5-mini model

You'll develop back-and-forth conversation based messaging with LLM and user where the AI maintains context across multiple user turns.  We will use Chat Streaming features when developing AI Chat Application.

Function Calling with .NET and gpt-5-mini model

Develop a function that will trigger from OpenAI GPT-5-mini. The model returns structured JSON specifying which .NET function to invoke, along with arguments for retrieving real-time data.

.NET AI Vector Search using Vector Embeddings and Vector Store

We’ll also cover Vector Search, a powerful feature that allows semantic search based on meaning—not keywords.

You’ll learn how to:

• Generate embeddings using OpenAI’s text-embedding-3-small or Ollama’s all-MiniLM  embeddings model,

• Store these in a vector database like Qdrant

• Query the vector store with user embedding to find top matches by similarity

• Retrieve relevant data based on similarity searches—all in our .NET applications.

RAG – Retrieval-Augmented Generation with .NET

You’ll learn how to combine vector search results with LLM responses to:

• Retrieve relevant data from your own sources

• Break documents into chunks → embed them → store in vector DB

• At query time, embed the question → retrieve relevant chunks → pass them along with the user’s query to the LLM

• Get accurate, context-specific answers using your internal data from LLM

We’ll implement the full RAG flow with real examples using .NET and Qdrant.

Image Analysis with .NET AI

Cover image recognition and analysis, showing how to send images to AI models, receive tags, captions or visual summaries and integrate those capabilities directly into your .NET apps

• Vision models for object recognition, classification, or captioning

• Combining text and image processing to build more powerful, multi-modal applications for traffic cam analysis operations

Final Project: E-Shop Semantic Search with .NET Aspire

You’ll build a complete full-stack AI-powered EShop Vector Search app step by step.

We’ll use:

• .NET Aspire for service orchestration

• Qdrant as our Vector Database

• and GPT-5 Mini or Ollama’s local models to generate embeddings and respond intelligently to user queries

In this project, you’ll:

• Generate product embeddings with OpenAI text-embeddings or Ollama all-minilm

• Store them in Qdrant Vector DB  for fast similarity search

• Implement a RAG flow that provides semantic search over our EShop product catalog

• Enable users to search products by meaning—not just keywords

This project brings everything you learn in this course into a single, full-stack, real-world app.

By the end of this course, you'll have the tools and confidence to build intelligent, GenAI-powered apps in .NET.