Artificial Intelligence
From symbolic AI and statistical learning to deep networks, transformers, LLMs and production model operations.
0 of 18 chapters passed
Pass mark is 75% on the chapter quiz.
1 · Origins & foundations
How the field began and the maths it runs on.
2 · Classical machine learning
Supervised and unsupervised learning that still runs most production systems.
- 3. The Supervised Learning WorkflowFeatures, labels, train/validation/test splits and the bias–variance trade-off.9 min · 3 MCQs
- 4. Core Algorithms: Regression to Gradient BoostingLinear and logistic regression, trees, ensembles and why boosting still wins on tabular data.11 min · 3 MCQs
- 5. Unsupervised Learning & RepresentationsClustering, dimensionality reduction and embeddings as the bridge to deep learning.9 min · 3 MCQs
- 6. Evaluating Models HonestlyAccuracy traps, precision/recall, ROC-AUC, calibration and fairness checks.9 min · 3 MCQs
3 · Deep learning
Neural networks, training dynamics and the vision/sequence architectures.
- 7. Neural Networks from First PrinciplesPerceptrons, activations, forward and backward passes, and what depth buys you.10 min · 3 MCQs
- 8. Training Dynamics That Actually MatterOptimisers, learning-rate schedules, normalisation, regularisation and mixed precision.10 min · 3 MCQs
- 9. Computer Vision & Convolutional NetworksConvolutions, the ImageNet moment, modern backbones and vision transformers.9 min · 3 MCQs
4 · Transformers & LLMs
Attention, pretraining, alignment and how modern language models behave.
- 10. Sequences: RNNs, LSTMs and the Attention BreakthroughWhy recurrence hit a ceiling and how attention removed it.10 min · 3 MCQs
- 11. Inside the TransformerTokenisation, embeddings, self-attention, multi-head projections and the decoder stack.12 min · 3 MCQs
- 12. How Large Language Models Are BuiltPretraining, scaling laws, instruction tuning, RLHF/DPO and the resulting behaviour.11 min · 3 MCQs
- 13. Inference, Decoding and HallucinationSampling parameters, KV caching, quantisation and why models state falsehoods confidently.10 min · 3 MCQs
5 · Applied & production AI
Prompting, fine-tuning, RAG, evaluation, serving and safety.
- 14. Prompt Engineering as a DisciplineStructure, few-shot examples, chain-of-thought, structured output and prompt injection.9 min · 3 MCQs
- 15. Fine-Tuning, LoRA and When Not ToFull fine-tuning vs parameter-efficient methods, dataset design and the decision tree.10 min · 3 MCQs
- 16. RAG and Agentic SystemsChunking, vector search, hybrid retrieval, reranking, tool use and agent loops.11 min · 3 MCQs
- 17. MLOps: Serving, Monitoring and DriftFrom notebook to endpoint: packaging, registries, CI/CD, observability and retraining.10 min · 3 MCQs
- 18. Safety, Ethics and GovernanceBias, privacy, evaluation harnesses, red-teaming and the regulatory landscape.9 min · 3 MCQs