AI Course

Lesson 1 • Core AI Subfields Overview

Maps the landscape of machine learning, computer vision, NLP, and robotics. Helps learners identify which subfield applies to their use cases.

Lesson 2 • How Machines Learn from Data

Explains supervised, unsupervised, and reinforcement learning at a conceptual level. Connects learning paradigms to real-world problem types.

Lesson 3 • Brief History of AI Development

Traces AI from symbolic logic to modern neural networks. Provides context for why current techniques dominate the field.

Lesson 4 • What AI Is and Is Not

Clarifies common misconceptions and establishes precise definitions of AI, ML, and deep learning. Sets shared vocabulary used throughout the course.

Lesson 1 • Understanding Data Types and Structures

Covers structured, unstructured, and semi-structured data and their roles in AI pipelines. Establishes the data vocabulary needed for all subsequent chapters.

Lesson 2 • Data Collection and Sourcing

Examines methods for gathering data including surveys, APIs, web scraping, and sensors. Highlights trade-offs between data richness and collection cost.

Lesson 3 • Data Quality and Cleaning

Identifies common data quality issues such as missing values, duplicates, and outliers. Teaches systematic cleaning workflows that improve model reliability.

Lesson 4 • Data Bias and Fairness Awareness

Identifies sources of bias in datasets and their downstream effects on AI outputs. Prepares learners to flag and mitigate bias before model training.

Lesson 1 • Model Training and Optimisation

Explains the training loop, loss functions, and gradient descent optimisation. Builds intuition for why models improve and when they stagnate.

Lesson 2 • Supervised Learning Algorithms

Covers regression, decision trees, and support vector machines with intuitive explanations. Connects algorithm choice to data characteristics and business goals.

Lesson 3 • Overfitting, Underfitting, and Regularisation

Diagnoses bias-variance trade-off and applies regularisation to improve generalisation. Directly prepares learners for robust model deployment.

Lesson 4 • Unsupervised Learning Techniques

Explores clustering and dimensionality reduction for unlabelled datasets. Shows how these techniques reveal hidden structure in business data.

Lesson 1 • Backpropagation and Weight Updates

Explains how gradients flow backward to update weights during training. Connects mathematical intuition to practical training stability.

Lesson 2 • Transfer Learning and Pretrained Models

Shows how to use pretrained weights to accelerate training on new tasks. Reduces compute cost and data requirements for practical projects.

Lesson 3 • Neural Network Architecture Basics

Introduces neurons, layers, weights, and activation functions as building blocks. Establishes the structural vocabulary for all deep learning topics ahead.

Lesson 4 • Convolutional Neural Networks

Covers convolution, pooling, and feature map extraction for image tasks. Prepares learners to use CNNs for visual recognition problems.

Lesson 1 • Cloud AI Platforms and Services

Surveys managed AI services for training, inference, and data storage in the cloud. Reduces infrastructure burden and accelerates time-to-deployment.

Lesson 2 • Experiment Tracking and Versioning

Covers logging hyperparameters, metrics, and artefacts across training runs. Enables reproducibility and systematic comparison of model iterations.

Lesson 3 • Building an End-to-End ML Pipeline

Integrates data ingestion, preprocessing, training, evaluation, and deployment into one workflow. Demonstrates how individual tools connect into a production-ready system.

Lesson 4 • Core ML Frameworks and Libraries

Introduces leading open-source frameworks for building and training models. Compares their strengths to guide framework selection for different project types.

Lesson 1 • NLP Applications and Use Cases

Uses NLP for sentiment analysis, named entity recognition, summarisation, and translation. Demonstrates end-to-end pipeline construction for each task type.

Lesson 2 • Text Preprocessing and Representation

Covers tokenisation, stemming, stop-word removal, and vectorisation methods. Prepares raw text for downstream NLP model consumption.

Lesson 3 • Large Language Models in Practice

Examines how LLMs are pretrained and fine-tuned for downstream tasks. Connects model capabilities to practical text generation and classification use cases.

Lesson 4 • Word Embeddings and Semantic Meaning

Explains dense vector representations that capture semantic relationships between words. Enables models to generalise across synonyms and related concepts.

Lesson 1 • Continuous Integration for ML

Applies software CI practices to model training, testing, and validation pipelines. Ensures code and model quality gates before any deployment proceeds.

Lesson 2 • Model Monitoring and Drift Detection

Tracks prediction quality, data drift, and concept drift in live systems. Enables proactive intervention before model degradation impacts business outcomes.

Lesson 3 • Model Governance and Documentation

Establishes model cards, audit trails, and approval workflows for responsible deployment. Connects technical deployment to organisational accountability standards.

Lesson 4 • Model Packaging and Serving

Covers serialisation formats, containerisation, and REST API creation for model serving. Bridges the gap between a trained model and a live application.

Lesson 1 • Building an AI Strategy for Organisations

Guides prioritisation of AI use cases, capability building, and ROI measurement. Equips learners to champion AI adoption at a strategic leadership level.

Lesson 2 • AI Risk Assessment and Management

Identifies technical, operational, and reputational risks in AI deployments. Teaches structured risk scoring and mitigation planning for AI initiatives.

Lesson 3 • AI Ethics Principles and Frameworks

Surveys fairness, accountability, transparency, and privacy as core ethical pillars. Provides a decision framework applicable to any AI project or product.

Lesson 4 • Regulatory and Compliance Landscape

Examines data protection obligations, algorithmic accountability requirements, and sector-specific AI rules. Prepares learners to navigate compliance without legal expertise.