Complete AI & LLM engineering course: ML foundations, scikit-learn, PyTorch, transformers, how LLMs work, prompt engineering, the Claude API, embeddings and vector search, RAG, agents and MCP, and shipping AI to production.
Each lecture has detailed notes, code examples, interview questions and practice exercises. Follow them in order, or jump to the topic you need.
Start your AI engineering journey: what artificial intelligence is, AI vs machine learning vs deep learning vs generative AI, supervised, unsupervised, reinforcement and self-supervised learning, the modern AI engineer role, and setting up Python, Jupyter, VS Code and Google Colab.
Learn the math and data foundations for machine learning: vectors, dot products and matrices, derivatives and gradient descent, mean, variance, distributions and Bayes' theorem, plus train/validation/test splits, data cleaning, feature scaling and categorical encoding with NumPy, pandas and scikit-learn.
Learn supervised machine learning with scikit-learn: the ML workflow, linear and logistic regression, KNN, decision trees, random forests and gradient boosting, pipelines, metrics like RMSE, R2, precision, recall, F1 and ROC-AUC, class imbalance, and saving models with joblib.
Learn unsupervised learning with k-means, DBSCAN and PCA, anomaly detection, feature engineering, the bias-variance trade-off, regularization, cross-validation, GridSearchCV and RandomizedSearchCV hyperparameter tuning, and how to avoid data leakage in scikit-learn.
Learn neural networks and deep learning with PyTorch from scratch: neurons, activation functions, loss functions, backpropagation, SGD and Adam optimizers, tensors, autograd, nn.Module, DataLoader, a full GPU training loop, CNNs for images, transfer learning with pretrained ResNet, and saving and loading models.
Learn NLP from text preprocessing and tokenization (words, subwords, BPE) to bag-of-words, TF-IDF and word embeddings, RNN/LSTM limits, attention and self-attention, the transformer architecture (encoder, decoder, positional encoding, multi-head attention), BERT vs GPT, and Hugging Face pipelines.
Understand how large language models work: next-token prediction, pretraining, tokens and context windows, temperature, top-k and top-p sampling, instruction tuning, RLHF, hallucination, knowledge cutoff, extended thinking, open-weight vs API models, and when to fine-tune, prompt or use RAG.
Learn prompt engineering for developers: the anatomy of a prompt, system prompts, few-shot examples, XML tags, chain-of-thought, structured JSON outputs, prompt chaining, reducing hallucinations, prompt evals and prompt injection defence, with runnable Claude API code in Python and Next.js.