AI Learning Course

Lesson 1 • Defining AI and Its Scope

Establishes precise definitions of AI, machine learning, and deep learning. Clarifies boundaries between narrow AI and general AI to prevent common misconceptions.

Lesson 2 • Core AI Paradigms and Approaches

Compares supervised, unsupervised, and reinforcement learning paradigms. Helps learners select the right approach for a given problem type.

Lesson 3 • History and Evolution of AI

Traces AI development from symbolic reasoning to modern neural networks. Contextualises current capabilities within decades of research breakthroughs and setbacks.

Lesson 1 • Bias in Data and Its Consequences

Identifies sampling, labelling, and historical bias types and their downstream effects. Equips learners to audit datasets before model training begins.

Lesson 2 • Data Collection and Sourcing

Examines methods for gathering training data from primary and secondary sources. Highlights how sourcing decisions affect model generalisability.

Lesson 3 • Exploratory Data Analysis for AI

Applies statistical summaries and visualisations to reveal data patterns and anomalies. Builds the analytical habit of understanding data before modelling.

Lesson 4 • Data Quality and Cleaning

Teaches identification and remediation of missing values, duplicates, and outliers. Directly prepares learners to deliver clean inputs to AI pipelines.

Lesson 1 • The Machine Learning Workflow

Maps the end-to-end process from problem framing to model deployment. Provides a repeatable framework applied throughout the rest of the course.

Lesson 2 • Supervised Learning Algorithms

Covers regression, decision trees, and ensemble methods with practical use cases. Builds algorithm intuition needed to choose models for real tasks.

Lesson 3 • Overfitting, Underfitting, and Regularisation

Diagnoses bias-variance trade-off and applies regularisation to improve generalisation. Directly addresses the most common failure mode in ML projects.

Lesson 4 • Unsupervised Learning Techniques

Introduces clustering, dimensionality reduction, and anomaly detection methods. Extends learner capability to problems without labelled training data.

Lesson 1 • Training Neural Networks

Covers backpropagation, gradient descent, and loss functions in practical terms. Connects mathematical intuition to hands-on training decisions.

Lesson 2 • Hyperparameter Tuning and Optimisation

Applies systematic search strategies to improve deep learning model performance. Bridges theoretical architecture knowledge to practical tuning workflows.

Lesson 3 • Recurrent Networks and Sequence Models

Covers RNNs, LSTMs, and GRUs for sequential and time-series data. Prepares learners for language and temporal modelling tasks.

Lesson 4 • Anatomy of a Neural Network

Explains neurons, layers, weights, and activation functions as building blocks. Establishes the vocabulary needed for all subsequent deep learning topics.

Lesson 1 • Text Preprocessing and Representation

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

Lesson 2 • Core NLP Tasks and Applications

Applies NLP to classification, named entity recognition, and summarisation tasks. Connects architecture knowledge to concrete business use cases.

Lesson 3 • Transformer Architecture and Attention

Explains self-attention, positional encoding, and the transformer block structure. Provides the architectural foundation for understanding large language models.

Lesson 4 • Fine-Tuning Pretrained Language Models

Demonstrates how to adapt pretrained models to domain-specific NLP tasks. Enables learners to leverage large models without training from scratch.

Lesson 1 • Generative Image and Multimodal Models

Covers diffusion models, GANs, and vision-language models for content generation. Expands learner capability beyond text to visual and multimodal outputs.

Lesson 2 • Retrieval-Augmented Generation

Combines external knowledge retrieval with LLM generation to reduce hallucination. Prepares learners to build grounded, factual AI applications.

Lesson 3 • How Large Language Models Work

Explains pretraining, tokenisation, and autoregressive generation in LLMs. Demystifies the mechanics behind conversational and generative AI outputs.

Lesson 4 • Evaluating and Auditing Generative Outputs

Applies factuality, coherence, and safety metrics to assess generative AI quality. Builds critical evaluation habits essential for responsible deployment.

Lesson 1 • Fairness Definitions and Measurement

Compares demographic parity, equalised odds, and individual fairness definitions. Enables learners to select and implement appropriate fairness criteria.

Lesson 2 • Privacy, Security, and Data Rights

Addresses differential privacy, data minimisation, and adversarial attack risks. Prepares learners to protect individuals and systems in AI deployments.

Lesson 3 • AI Governance and Policy Frameworks

Examines risk-tiered governance models, audit processes, and accountability structures. Equips learners to design or comply with organisational AI governance.

Lesson 4 • Transparency and Explainability

Covers interpretable models, SHAP, LIME, and model cards for AI transparency. Connects explainability to stakeholder trust and accountability requirements.

Lesson 1 • Scaling AI Infrastructure

Examines cloud compute, distributed training, and hardware acceleration options. Connects infrastructure decisions to cost, speed, and scalability goals.

Lesson 2 • Model Packaging and Serving

Covers containerisation, REST APIs, and batch vs. real-time serving architectures. Prepares learners to expose models as reliable, scalable services.

Lesson 3 • MLOps Principles and Pipelines

Introduces MLOps as the discipline bridging ML development and production operations. Establishes pipeline thinking as the foundation for reliable AI delivery.

Lesson 4 • AI Project Lifecycle Management

Applies agile and iterative methods to manage AI projects from ideation to retirement. Integrates technical deployment skills with project governance practices.