Microsoft AI training

Lesson 1 • What AI Is and How It Works

Covers core AI definitions, the difference between AI, ML, and deep learning, and real-world analogies. Establishes shared vocabulary used throughout the course.

Lesson 2 • History and Evolution of AI

Traces AI from early symbolic systems to modern neural networks. Provides context for why current Microsoft tools exist and how they evolved.

Lesson 3 • Key AI Concepts and Terminology

Defines training data, models, inference, tokens, and embeddings. These terms recur in every subsequent chapter and must be internalized early.

Lesson 4 • Responsible AI Principles

Introduces Microsoft's six responsible AI principles and their practical implications. Sets the ethical baseline applied in all later applied chapters.

Lesson 1 • Azure Language and Text Services

Explores sentiment analysis, entity recognition, key phrase extraction, and translation. Connects text AI capabilities to document processing and customer feedback scenarios.

Lesson 2 • Azure Computer Vision Services

Covers image analysis, OCR, face detection, and custom vision models. Students integrate vision APIs into sample workflows to extract structured data from images.

Lesson 3 • Azure AI Search and Knowledge Mining

Introduces cognitive search, skillsets, and knowledge stores for extracting insights from unstructured data. Bridges raw data ingestion to intelligent retrieval.

Lesson 4 • Azure Speech Services

Covers speech-to-text, text-to-speech, and speaker recognition APIs. Students build a basic voice-enabled interaction using Azure Speech SDK.

Lesson 1 • Monitoring and Retraining Models

Introduces data drift detection, model performance monitoring, and automated retraining triggers. Ensures deployed models remain accurate over time.

Lesson 2 • Preparing and Managing Data

Teaches data ingestion, profiling, cleaning, and dataset versioning within Azure ML. Clean, versioned data is the prerequisite for reproducible model training.

Lesson 3 • Azure Machine Learning Workspace Setup

Covers workspace creation, compute targets, datastores, and the studio interface. Establishes the environment used for all ML experiments in this chapter.

Lesson 4 • Model Evaluation and Deployment

Covers evaluation metrics, model registration, and real-time or batch endpoint deployment. Students publish a working REST endpoint and test it with sample data.

Lesson 5 • Training Models with Automated ML

Uses AutoML to run classification, regression, and forecasting experiments automatically. Students interpret AutoML results and select the best-performing model.

Lesson 1 • Accessing Azure OpenAI Service

Covers provisioning, model deployment, and API authentication in Azure OpenAI. Students make their first API call and inspect request and response structures.

Lesson 2 • Fine-Tuning and Customization

Covers fine-tuning workflows, training file preparation, and evaluating fine-tuned models. Students understand when fine-tuning outperforms prompt engineering alone.

Lesson 3 • Introduction to Large Language Models

Explains transformer architecture, tokenization, and how LLMs generate text. Provides the conceptual grounding needed to use Azure OpenAI effectively.

Lesson 4 • Prompt Engineering Techniques

Teaches zero-shot, few-shot, chain-of-thought, and system prompt strategies. Effective prompting directly determines output quality for all downstream applications.

Lesson 1 • Copilot in Microsoft 365 Apps

Demonstrates Copilot features in Word, Excel, PowerPoint, Outlook, and Teams. Students practice prompts that automate drafting, summarization, and data analysis tasks.

Lesson 2 • Extending Copilot with Plugins

Covers plugin development, connector integration, and API action configuration in Copilot Studio. Plugins allow Copilot to retrieve live data and execute external actions.

Lesson 3 • Microsoft Copilot Architecture

Explains how Copilot connects Microsoft Graph, LLMs, and app context to generate responses. Understanding the architecture helps users craft better prompts and set expectations.

Lesson 4 • Microsoft Copilot Studio Basics

Introduces the low-code environment for building custom Copilot agents and topics. Students create a functional agent that handles a defined business question.

Lesson 1 • AI Builder Overview and Models

Surveys prebuilt and custom AI Builder models including form processing, prediction, and object detection. Students identify which model type fits a given automation scenario.

Lesson 2 • Prediction Models in Power Apps

Builds a binary prediction model and surfaces results inside a canvas Power App. Students connect model output to app logic for data-driven decision support.

Lesson 3 • Copilot in Power Platform

Demonstrates natural language app and flow creation using Copilot inside Power Apps and Power Automate. Students build a functional app by describing requirements in plain language.

Lesson 4 • Form Processing and Document AI

Trains a form processing model to extract fields from invoices or contracts. Connects directly to Power Automate for automated document routing and data entry.

Lesson 1 • Azure AI Studio Environment

Introduces the hub, project, and resource model in Azure AI Studio. Students configure a project connected to Azure OpenAI, AI Search, and storage resources.

Lesson 2 • Deploying and Monitoring AI Apps

Deploys a Prompt Flow solution as a managed online endpoint and configures monitoring. Connects deployment practices from earlier chapters to a production-ready workflow.

Lesson 3 • Prompt Flow Development

Builds orchestration flows using Prompt Flow nodes for LLM calls, Python tools, and search. Students create a multi-step flow that processes user input end to end.

Lesson 4 • Model Catalog and Selection

Explores the model catalog including OpenAI, Meta, Mistral, and Hugging Face models. Students apply selection criteria based on task type, cost, and latency requirements.

Lesson 1 • Data Strategy for AI at Scale

Addresses data governance, data mesh concepts, and enterprise data platform design for AI. A sound data strategy is the prerequisite for scalable, trustworthy AI systems.

Lesson 2 • Building an AI Center of Excellence

Defines the roles, operating model, and governance structure of an AI Center of Excellence. Students draft a CoE charter aligned to their organization's AI maturity level.

Lesson 3 • AI Governance Frameworks

Covers model risk management, audit trails, explainability requirements, and policy structures. Governance frameworks protect organizations from regulatory and reputational risk.

Lesson 4 • AI Change Management and Adoption

Covers stakeholder engagement, workforce upskilling, and resistance management for AI rollouts. Organizational adoption determines whether technically sound AI delivers business value.