Azure AI Course

Lesson 1 • Azure AI Services Landscape

Maps the full catalogue of Azure AI services across vision, language, speech, and decision domains. Provides context for selecting the right service for a given business problem.

Lesson 2 • Azure Portal and CLI Navigation

Hands-on orientation to the Azure portal, Azure CLI, and Azure PowerShell. Students gain the operational fluency required to deploy and manage AI resources.

Lesson 3 • Cloud Computing Essentials for AI

Covers IaaS, PaaS, and SaaS models and how Azure structures compute, storage, and networking. Establishes the cloud vocabulary needed throughout the course.

Lesson 1 • Pipelines and Workflow Automation

Introduces Azure ML pipelines for chaining data prep, training, and evaluation steps. Automation reduces manual errors and enables repeatable production workflows.

Lesson 2 • Data Ingestion and Preparation

Teaches loading, profiling, and transforming datasets within Azure ML. Clean, well-structured data is the prerequisite for all subsequent model training.

Lesson 3 • Azure ML Workspace Setup

Covers workspace creation, compute targets, and associated storage accounts. Connects portal skills from Chapter 1 to the ML-specific environment.

Lesson 4 • Training Experiments and Tracking

Demonstrates submitting training runs, logging metrics, and comparing experiments in MLflow. Builds the habit of reproducible, auditable ML workflows.

Lesson 1 • Featurisation and Data Guardrails

Covers automatic featurisation, imputation strategies, and data guardrail warnings. Students learn to interpret and override AutoML preprocessing decisions.

Lesson 2 • Responsible AutoML Practices

Addresses bias detection, fairness metrics, and model explainability within AutoML outputs. Aligns automated model selection with organisational AI governance standards.

Lesson 3 • AutoML Configuration and Task Types

Explains how to configure AutoML runs for different task types and set exit criteria. Builds on dataset registration skills from Chapter 2.

Lesson 1 • Model Retraining and CI/CD Pipelines

Covers triggering retraining on data drift and automating deployment with Azure DevOps. Closes the MLOps loop from training through production monitoring.

Lesson 2 • Endpoint Monitoring and Logging

Demonstrates enabling Application Insights, collecting inference logs, and setting latency alerts. Monitoring ensures deployed models meet service-level expectations.

Lesson 3 • Containerisation with Docker and ACR

Introduces packaging models into Docker containers and pushing images to Azure Container Registry. Enables portable, environment-consistent deployments.

Lesson 4 • Real-Time Endpoint Deployment

Covers deploying models to managed online endpoints with scoring scripts and environments. Builds directly on model registration skills from Chapter 2.

Lesson 1 • Face API and Spatial Analysis

Introduces face detection, attribute analysis, and responsible use constraints of the Face API. Connects to ethical AI principles established in Chapter 3.

Lesson 2 • Speech Translation and Speaker Recognition

Demonstrates real-time speech translation across languages and speaker identification. Expands speech capabilities for multilingual and multi-speaker scenarios.

Lesson 3 • Custom Vision for Image Classification

Teaches training and publishing custom image classifiers and object detectors in Custom Vision. Extends pre-built APIs to domain-specific visual recognition tasks.

Lesson 4 • Computer Vision API Fundamentals

Covers image analysis, object detection, and OCR using the Computer Vision API. Introduces the REST call pattern used across all Cognitive Services.

Lesson 1 • Document Intelligence and Form Recognizer

Teaches extracting structured data from forms, invoices, and receipts using Document Intelligence. Automates document processing workflows in business operations.

Lesson 2 • Text Analytics and Sentiment Analysis

Covers sentiment scoring, opinion mining, and key phrase extraction via the Language API. Provides the NLP baseline for all subsequent language tasks in this chapter.

Lesson 3 • Question Answering with Azure AI

Covers building knowledge bases from documents and FAQs using the Question Answering service. Enables self-service information retrieval in enterprise applications.

Lesson 4 • Named Entity Recognition and Linking

Teaches identifying and classifying entities in text and linking them to knowledge bases. Enables structured information extraction from unstructured documents.

Lesson 1 • Building Generative AI Applications

Demonstrates integrating Azure OpenAI into web apps using the REST API and SDKs. Students build a functional chat application with conversation history management.

Lesson 2 • Fine-Tuning and Custom Models

Covers preparing training data, submitting fine-tuning jobs, and evaluating custom model quality. Extends base models for domain-specific language and style requirements.

Lesson 3 • Prompt Engineering Fundamentals

Covers zero-shot, few-shot, and chain-of-thought prompting techniques with practical examples. Effective prompts are the primary lever for controlling generative model output.

Lesson 4 • Embeddings and Semantic Search

Teaches generating text embeddings and building vector similarity search pipelines. Enables retrieval-augmented generation and semantic document search applications.

Lesson 1 • AI Governance and Compliance Frameworks

Covers Microsoft's Responsible AI standard, model cards, and audit logging for compliance. Aligns solution design with organisational and regulatory accountability requirements.

Lesson 2 • Cost Optimisation Strategies

Teaches right-sizing compute, using spot instances, and monitoring AI service consumption costs. Enables teams to deliver AI solutions within budget constraints.

Lesson 3 • Networking and Private Connectivity

Covers virtual networks, private endpoints, and service firewalls for AI resource isolation. Ensures AI workloads meet enterprise network security requirements.

Lesson 4 • Operationalising AI at Enterprise Scale

Addresses MLOps maturity models, team structures, and platform governance for large organisations. Prepares students to lead AI adoption beyond individual project delivery.