Machine Learning
11 Modules ~40 hours Beginner → Advanced
Master Machine Learning end-to-end: math foundations, supervised/unsupervised algorithms, feature engineering, evaluation, neural networks, and ML in production. Builds toward our MLflow and llm-development tracks.
Course roadmap
| # | Module | Status | Topics |
|---|---|---|---|
| 0 | Setup & Math Refresher | Plan ready | Python, numpy, pandas, linear algebra essentials, calculus for grads |
| 1 | ML Fundamentals | Plan ready | Supervised vs unsupervised vs RL, train/val/test split, the bias-variance tradeoff |
| 2 | Linear Models | Plan ready | Linear/logistic regression, regularization (L1/L2), gradient descent |
| 3 | Tree-Based Models | Plan ready | Decision trees, random forests, gradient boosting (XGBoost, LightGBM) |
| 4 | Unsupervised Learning | Plan ready | k-means, DBSCAN, hierarchical clustering, PCA, t-SNE, UMAP |
| 5 | Feature Engineering | Plan ready | Encoding, scaling, target encoding, feature selection, leakage detection |
| 6 | Model Evaluation | Plan ready | Cross-validation, metrics (precision, recall, AUC, MAE), error analysis |
| 7 | Neural Networks | Plan ready | Perceptron → MLP → backprop, PyTorch basics, GPU usage |
| 8 | Deep Learning Architectures | Plan ready | CNN, RNN/LSTM, Transformers introduction |
| 9 | ML in Production | Plan ready | MLOps overview, model registry, monitoring, drift, fairness |
| 10 | Capstone | Plan ready | Build an end-to-end ML pipeline: data → model → registered → API |
What's available now
Curriculum plan published. Content rolling out 2026 H2.
Related courses:
- Python — language foundations
- MLflow — track and serve models (full course already)
- llm-development — LLM-specific techniques
Last updated
2026-05 — Curriculum plan published.