AI Engineer Course

Lesson 1 • Python for AI Development

Establishes Python proficiency focused on AI workflows using NumPy, Pandas, and Matplotlib. Provides the coding baseline required for all subsequent chapters.

Lesson 2 • Data Fundamentals for AI

Explains data types, structures, and quality requirements that drive model performance. Prepares engineers to evaluate datasets before modelling begins.

Lesson 3 • AI Landscape and Key Terminology

Defines AI, ML, and deep learning with precise distinctions. Establishes shared vocabulary used throughout the entire course.

Lesson 1 • Ensemble and Boosting Methods

Explains bagging, random forests, gradient boosting, and XGBoost for improved predictive power. Builds on single-algorithm knowledge to show ensemble advantages.

Lesson 2 • Unsupervised Learning Techniques

Teaches clustering, dimensionality reduction, and anomaly detection for unlabelled data. Expands the engineer's toolkit beyond supervised paradigms.

Lesson 3 • Feature Engineering and Selection

Covers encoding, scaling, interaction features, and selection methods that directly improve model accuracy. Bridges raw data preparation to algorithm input requirements.

Lesson 4 • Model Evaluation and Selection

Defines metrics, cross-validation, and bias-variance trade-off for rigorous model assessment. Ensures engineers can justify model choices with quantitative evidence.

Lesson 1 • Recurrent and Sequence Models

Explains RNNs, LSTMs, and GRUs for sequential and time-series data modelling. Prepares engineers for temporal patterns before transformer architectures are introduced.

Lesson 2 • Neural Network Fundamentals

Explains perceptrons, activation functions, forward propagation, and backpropagation mathematically. Grounds deep learning in the ML foundations established previously.

Lesson 3 • Convolutional Neural Networks

Teaches CNN architecture, pooling, and transfer learning for image and spatial data tasks. Extends neural network knowledge to computer vision applications.

Lesson 4 • Training Deep Networks Effectively

Covers optimisers, learning rate schedules, regularisation, and batch normalisation for stable training. Addresses common failure modes engineers encounter in practice.

Lesson 1 • Large Language Model Engineering

Covers LLM APIs, prompt engineering, context windows, and output parsing for production use. Equips engineers to integrate LLMs reliably into real applications.

Lesson 2 • Classical NLP Techniques

Covers tokenisation, stemming, TF-IDF, and word embeddings as NLP building blocks. Establishes the baseline before transformer-based methods are introduced.

Lesson 3 • Transformer Architecture Deep Dive

Explains self-attention, multi-head attention, positional encoding, and encoder-decoder design. Provides the architectural understanding needed to work with modern LLMs.

Lesson 4 • Retrieval-Augmented Generation Systems

Builds RAG pipelines combining vector search with LLM generation for knowledge-grounded responses. Extends LLM usage to enterprise document and knowledge base scenarios.

Lesson 1 • AI System Design Principles

Introduces design patterns, trade-off analysis, and system thinking specific to AI workloads. Frames architecture decisions within the constraints of production environments.

Lesson 2 • Scalability and Infrastructure for AI

Covers containerisation, orchestration, and cloud infrastructure patterns for AI workloads at scale. Prepares engineers to deploy systems that grow with demand.

Lesson 3 • Data Pipelines and Feature Stores

Covers ETL design, streaming ingestion, and feature store architecture for consistent feature delivery. Ensures training and serving features remain synchronised.

Lesson 4 • Designing for Reliability and Observability

Teaches circuit breakers, fallback strategies, logging, and alerting for AI service resilience. Ensures engineers can maintain system health in production.

Lesson 1 • CI/CD for Machine Learning

Applies continuous integration and delivery practices to ML code, data, and model artefacts. Bridges software engineering discipline with ML development workflows.

Lesson 2 • Model Monitoring and Drift Detection

Implements data drift, concept drift, and performance monitoring to detect model degradation. Enables proactive maintenance before model failures impact users.

Lesson 3 • ML Pipelines and Workflow Orchestration

Designs automated ML pipelines using orchestration tools for training, validation, and deployment. Reduces manual intervention and accelerates iteration cycles.

Lesson 4 • Experiment Tracking and Reproducibility

Covers experiment logging, artefact versioning, and environment management for reproducible ML. Establishes the foundation for disciplined model development.

Lesson 1 • Responsible AI Governance Frameworks

Applies responsible AI principles, impact assessments, and policy alignment to engineering decisions. Connects technical controls to organisational accountability structures.

Lesson 2 • Bias, Fairness, and Equity in AI

Identifies sources of algorithmic bias and applies fairness metrics to detect and mitigate disparate impact. Grounds ethical reasoning in measurable engineering actions.

Lesson 3 • AI Safety and Risk Management

Teaches threat modelling, adversarial robustness, and failure mode analysis for AI systems. Prepares engineers to anticipate and mitigate harmful AI behaviours.

Lesson 4 • Privacy-Preserving AI Techniques

Covers differential privacy, federated learning, and data anonymisation for privacy-safe AI. Addresses data protection requirements without sacrificing model utility.

Lesson 1 • Project Review and Presentation

Presents the end-to-end system with architecture diagrams, results, and lessons learned. Develops communication skills needed to deliver AI projects to technical audiences.

Lesson 2 • Problem Framing and Scoping

Translates a business problem into a well-defined AI task with success metrics and constraints. Applies system design thinking from earlier chapters to a real scenario.

Lesson 3 • Data Acquisition and Preparation

Executes data collection, cleaning, and feature engineering for the capstone dataset. Applies all data skills from previous chapters in an integrated workflow.

Lesson 4 • Model Development and Experimentation

Runs tracked experiments, selects the best model, and documents decisions with justification. Demonstrates disciplined ML development using MLOps practices.