AI Fundamentals Course

Lesson 1 • Societal and Economic Impact

Examines how AI reshapes labour markets, productivity, and social structures. Prepares learners to engage critically with AI's broader consequences.

Lesson 2 • Defining Artificial Intelligence

Establishes a precise, working definition of AI distinct from related terms. Anchors all subsequent learning in shared vocabulary.

Lesson 3 • A Brief History of AI

Traces AI from early symbolic systems to modern neural networks. Provides context for understanding why current approaches dominate.

Lesson 1 • How Machines Learn from Data

Explains the learning loop: data input, model training, and prediction output. Sets the conceptual stage for all ML algorithm discussions.

Lesson 2 • Supervised Learning Fundamentals

Covers regression and classification as the two primary supervised tasks. Learners connect algorithm choice to problem type.

Lesson 3 • Bias, Variance, and Model Fit

Explains underfitting and overfitting as core model quality problems. Equips learners to diagnose and discuss model performance issues.

Lesson 4 • Unsupervised Learning Fundamentals

Introduces clustering and dimensionality reduction for unlabeled data. Expands learners' toolkit beyond labeled-data scenarios.

Lesson 1 • Data Quality and Cleaning

Addresses missing values, duplicates, outliers, and inconsistencies as quality threats. Learners apply cleaning strategies to improve downstream model reliability.

Lesson 2 • Data Collection and Sourcing

Covers primary collection, open datasets, and synthetic data generation methods. Helps learners identify appropriate data sources for AI projects.

Lesson 3 • Types of Data in AI

Categorises structured, unstructured, and semi-structured data with examples. Grounds data strategy decisions in concrete data type awareness.

Lesson 4 • Feature Engineering Essentials

Transforms raw variables into informative model inputs through encoding and scaling. Directly improves model performance by enriching input representations.

Lesson 1 • Training with Backpropagation

Explains gradient descent and backpropagation as the core training mechanism. Demystifies how networks improve through iterative weight updates.

Lesson 2 • Feedforward Neural Networks

Describes multi-layer networks, forward propagation, and loss computation. Learners trace data flow from input to prediction.

Lesson 3 • Biological Inspiration and Perceptrons

Connects neuron biology to the mathematical perceptron model. Establishes the building block for all neural network architectures.

Lesson 4 • Convolutional Neural Networks

Introduces CNNs as specialised architectures for image and spatial data. Connects filter operations to feature detection in visual tasks.

Lesson 1 • Text Representation Techniques

Converts raw text into numerical formats machines can process. Establishes the input layer for all NLP model pipelines.

Lesson 2 • Conversational AI and Chatbots

Covers dialogue systems, intent recognition, and retrieval-augmented generation. Connects NLP theory to deployed conversational product architectures.

Lesson 3 • Core NLP Tasks and Applications

Surveys sentiment analysis, named entity recognition, and text classification. Maps NLP capabilities to practical business and research use cases.

Lesson 4 • The Transformer Architecture

Explains attention mechanisms and the transformer design that powers modern NLP. Provides conceptual grounding for understanding large language models.

Lesson 1 • Transparency and Explainability

Distinguishes interpretable models from black-box systems and introduces XAI tools. Connects explainability to stakeholder trust and regulatory compliance.

Lesson 2 • Sources of Bias in AI Systems

Traces bias origins from data collection through model deployment. Enables learners to audit AI pipelines for discriminatory patterns.

Lesson 3 • Responsible AI Governance Frameworks

Surveys organisational and cross-sector frameworks for accountable AI development. Prepares learners to implement governance structures in their organisations.

Lesson 4 • Fairness Metrics and Definitions

Introduces statistical fairness criteria including demographic parity and equalised odds. Learners compare definitions and understand their inherent tradeoffs.

Lesson 1 • Evaluation Metrics in Depth

Covers accuracy, precision, recall, F1, AUC-ROC, and regression metrics. Learners select and interpret metrics appropriate to each problem context.

Lesson 2 • Framing the AI Problem

Translates business or research questions into well-defined ML problem statements. Prevents costly misalignment between stakeholder goals and model design.

Lesson 3 • Model Selection and Comparison

Guides algorithm selection based on data characteristics and performance requirements. Learners apply structured comparison to choose the best model candidate.

Lesson 4 • Experiment Tracking and Reproducibility

Introduces logging, versioning, and documentation practices for ML experiments. Ensures learners can reproduce and communicate results reliably.

Lesson 1 • Monitoring and Model Drift

Establishes monitoring pipelines to detect data drift, concept drift, and performance decay. Enables proactive maintenance before model degradation impacts users.

Lesson 2 • MLOps Principles and Practices

Introduces CI/CD pipelines, automated testing, and model registry management for ML. Aligns AI development with modern software engineering reliability standards.

Lesson 3 • AI System Risk and Incident Management

Prepares teams to identify, respond to, and learn from AI system failures in production. Closes the chapter with a safety-first operational mindset.

Lesson 4 • Scalability and Infrastructure

Addresses load balancing, auto-scaling, and hardware acceleration for AI workloads. Prepares learners to design infrastructure that handles production traffic.