35 Problems

Cracking ML

The definitive set of 35 machine learning problems for ML engineer interviews. All NumPy from scratch, no sklearn/PyTorch. Covers linear models, classification, trees, unsupervised learning, evaluation, and feature engineering.

13 Easy16 Medium6 Hard

Unlock the full study plan

Upgrade to Pro to access the full curriculum.

Upgrade to Pro

6 sections

Linear Regression from Scratch Logistic Regression from Scratch Ridge Regression Lasso Regression Softmax Regression Linear Models Quiz10 questions

KNN Classifier Gaussian Naive Bayes Averaged Perceptron Linear Discriminant Analysis SVM with Hinge Loss (SGD)Classification Quiz10 questions

Permutation Feature Importance Decision Tree Classifier (CART)Decision Tree Regressor Bagging Classifier Random Forest from Scratch AdaBoost from Scratch Gradient Boosted Regressor Stacking Ensemble Ensemble Methods & Trees Quiz10 questions

K-Means Clustering K-Means++ Initialization DBSCAN Agglomerative Clustering PCA from Scratch Gaussian Mixture Models (EM)Isolation Forest t-SNE Affinity Matrix Unsupervised Learning Quiz10 questions

K-Fold Cross-Validation Log Loss (Binary Cross-Entropy)AUC-ROC from Scratch Precision-Recall Curve & Average Precision Regression Metrics Evaluation & Model Selection Quiz10 questions

Missing Value Imputation Categorical Encoding Feature Scaling Distance Metrics Feature Engineering & Preprocessing Quiz10 questions

Summary

Total35

Easy13

Medium16

Hard6

Progress0%

Quizzes0/6