Microsoft's AI courses serve several development stages: understanding machine learning, building a generative AI application, connecting agents with tools, and operating or evaluating the result. Choose a route that matches your programming background and intended system. A protocol or deployment course is most useful when you understand the application it supports.
Table of Contents
Courses to consider first
| Learning goal | Recommended course |
|---|---|
| For machine-learning foundations | Machine Learning Course: Learn the Fundamentals |
| For a generative AI application route | Generative AI Course: Build AI Applications |
| For intelligent agent applications | AI Agents Course: Build Intelligent Applications |
| For connecting tools and data | Model Context Protocol (MCP) Course |
For machine-learning foundations: Start with the fundamentals if models and their role in an application are still unfamiliar.
For a generative AI application route: The application-building course gives the broader developer context for the more specialist agent and retrieval topics.
For intelligent agent applications: Choose the AI Agents course when agent behaviour and application design are your main learning goal.
For connecting tools and data: Use the MCP course when you need to understand that integration layer rather than another general AI introduction.
19 courses in this Microsoft selection
-
1. AI Agents Course: Build Intelligent Applications
Move from the idea of an AI agent to the design decisions behind a working agent system. Microsoft introduces agent frameworks, tools, retrieval, planning, memory, and multi-agent collaboration through focused video lessons.
-
2. Java AI Course: Generative AI and MCP
Explore how generative AI fits into Java applications. The series starts with AI techniques for Java developers and extends into application development, MCP, and modernizing existing projects.
-
3. Model Context Protocol (MCP) Course
Learn the structure of Model Context Protocol and how it connects AI applications with tools and resources. The video series moves through core concepts, security, implementation, and examples of MCP in development workflows.
-
4. Generative AI Course: Build AI Applications
Build a foundation for creating applications with generative AI. Microsoft covers language models, comparing model options, responsible use, and prompt engineering before moving into application-focused lessons.
-
5. LLMOps Course: Deploy and Monitor AI Apps
Follow the lifecycle of an LLM application through five demonstrations: exploration, retrieval augmentation, code-based flows, evaluation, and production monitoring. The series uses Azure tooling to connect development decisions with operational checks.
-
6. Machine Learning Course: Learn the Fundamentals
Start with classical machine learning and the tools used to experiment with models. Microsoft introduces the field, its history and techniques, then demonstrates practical work with Python notebooks and common data libraries.
-
7. AI Agent Context Course: Foundry IQ and Work IQ
Understand how AI agents obtain useful context from knowledge sources and workplace tools. This six-part Microsoft series covers Foundry IQ data pipelines and knowledge-base queries, followed by Work IQ context, agent-to-agent interactions, and MCP or CLI tooling.
-
8. AI Toolkit Course: Build and Evaluate AI Agents
Build a clear path through Microsoft AI Toolkit and GitHub Copilot: prepare the environment, choose a model, add tools, generate agent code, inspect traces, and evaluate output.
-
9. Microsoft Foundry Course: AI Agent Development
Explore the practical steps involved in building and improving AI agents in Microsoft Foundry. The demonstrations include Python, VS Code, visual workflows, model selection, trusted tools, context, and troubleshooting.
-
10. AI Red Teaming Course: Prompt Injection and PyRIT
Learn how generative AI systems are tested for unsafe behavior through Microsoft’s ten-part AI Red Teaming 101 series. The course introduces model behavior, direct and indirect prompt injection, single-turn and multi-turn testing, defenses, and PyRIT automation.
-
11. Azure AI Course: Multimodal Content Understanding
Explore how Azure AI Content Understanding works with more than plain text. Three demonstrations cover getting started, using video in retrieval-augmented generation, and analyzing calls.
-
12. RAG Course: Retrieval, Vector Search and Multimodal Data
Study retrieval-augmented generation through five structured learning journeys. The course starts with knowledge retrieval, then examines retrieval system design, vector index scaling, multimodal data, and application scenarios.
-
13. Generative AI with JavaScript Course: LLMs and RAG
Learn the building blocks of generative AI applications in JavaScript. The series connects language models and prompt design with RAG, LangChain.js, local models through Ollama, Azure services, and streaming responses.
-
14. Responsible AI Course: Content Safety and Evaluation
Practice thinking about AI safety as a set of explicit checks. These demonstrations cover text and image moderation, blocklists, groundedness, prompt shields, bulk testing, content filters, and manual or automated evaluation in Azure AI tools.
-
15. Azure AI Language Course: NLP and Text Analysis
Explore natural language processing through short Azure AI Language demonstrations. The playlist covers summarization, assisted data labeling, call transcription analysis, translation, and personal information redaction.
-
16. Responsible AI Course: Image and Text Model Review
Follow a two-part introduction to responsible AI support for image and text models. This compact series is useful for learners considering how evaluation needs differ across model inputs.
-
17. MLOps Course: Azure Machine Learning Workflows
Explore MLOps and Azure Machine Learning through a three-part technical series. The lessons are presented in numbered order to support a connected discussion of scaling machine learning work.
-
18. Machine Learning with R Course: Tidymodels
Learn a connected introduction to machine learning in R with tidymodels. Four sessions cover data exploration, regression, classification, and clustering.
-
19. AI Fundamentals Course: Machine Learning Concepts
Build a conceptual starting point for artificial intelligence with David Carmona’s AI 101 series. The short whiteboard lessons introduce machine learning and classification through accessible explanations.
Choose your learning route
- Establish the relevant ML or generative AI concepts and the programming background needed for your route.
- Build application understanding before specializing in agents, RAG or MCP.
- Add evaluation, responsible AI, LLMOps or MLOps when you need to assess and operate the system.
Frequently asked questions
Which course should I choose if ML concepts are unfamiliar?
Machine Learning Fundamentals is the broad conceptual starting point in this selection. Agent and protocol courses address more specific development questions.
Should I study MCP before generative AI applications?
Application context helps explain why the protocol matters. Choose MCP when connecting an application to tools or data is the question you need to answer.
Questions & discussion
Share a useful question or correction. Comments appear after moderation. Please avoid personal or sensitive information.