Role-specific and topic-specific questions with answers
Easy, medium, and hard questions that cover most topics in machine learning and data science interviews. Solutions that deep dive into explaining complicated concepts, with necessary references and simulations when needed.
| Problem | Role | Area | Topics | Difficulty | Status | Overall rating |
|---|---|---|---|---|---|---|
| Normality assumption | PDS DS AS | A/B Testing | Normality | Medium | ||
| Reducing variance in AB testing | PDS DS AS | A/B Testing | Variance | Medium | ||
| Sample Ratio Mismatch | PDS DS AS | A/B Testing | Sample Ratio Mismatch | Medium | ||
| Simpson’s paradox | PDS DS AS | A/B Testing | Simpson’s paradox | Medium | ||
| Questions worksheet (recruiter-screen) | PDS DS AS MLE | Application process | Discussion prep, Application prep | Medium | ||
| Referrals vs. online applications | PDS DS AS MLE | Application process | Application prep | Easy | ||
| Response time | PDS DS AS MLE | Application process | Application prep | Easy | ||
| Timeline: from application to offer | PDS DS AS MLE | Application process | Application prep | Easy | ||
| Questions worksheet (behavioral) | PDS DS AS MLE | Behavioral | Discussion prep | Hard | ||
| Climbing stairs (Leetcode 70) | DS AS MLE | Data Structures and Algorithms | Recursion, Dynamic programming | Easy | ||
| Find if Path Exists in Graph (Leetcode 1971) | DS AS MLE | Data Structures and Algorithms | DFS, BFS | Medium | ||
| K Closest Points to Origin (Leetcode 973) | DS AS MLE | Data Structures and Algorithms | Heap | Easy | ||
| Longest substring without repeating characters (Leetcode 3) | DS AS MLE | Data Structures and Algorithms | Hash, Sliding Window | Medium | ||
| Minimum remove to make valid parentheses (Leetcode 1249) | DS AS MLE | Data Structures and Algorithms | Stack | Medium | ||
| Number of recent calls (Leetcode 933) | DS AS MLE | Data Structures and Algorithms | Queue, DEQueue | Easy | ||
| Remove duplicates in place (Leetcode 26) | DS AS MLE | Data Structures and Algorithms | Array, Two pointers | Easy | ||
| Search in Binary Search Tree (Leetcode 700) | DS AS MLE | Data Structures and Algorithms | Binary Search, Binary Search Tree | Easy | ||
| Sort an Array (Leetcode 912) | DS AS MLE | Data Structures and Algorithms | Recursion, Sorting | Medium | ||
| Analyze prediction error | DS AS MLE | Machine Learning | Prediction error, Bias-variance tradeoff, Diagnostics, Learning curves | Medium | ||
| Bias-variance equation | DS AS MLE | Machine Learning | Bias-variance tradeoff, Formula derivation | Easy | ||
| Linear regression with gradient descent | DS AS MLE | Machine Learning Coding | Gradient descent, Linear regression | Medium | ||
| Sample from random generator | DS AS MLE | Machine Learning Coding | Sample, Uniform, Random number generator | Medium | ||
| Gradient descent vs. stochastic gradient descent | DS AS MLE | Machine Learning | Gradient descent, Stochastic gradient descent, Minibatch | Medium | ||
| Imbalanced dataset | PDS DS AS MLE | Machine Learning | Class imbalance | Medium | ||
| k-means | PDS DS AS MLE | Machine Learning | k-means | Medium | ||
| L1 (Lasso) vs. L2 (Ridge) regularization | DS AS MLE | Machine Learning | L1, L2, Redularization, Lasso, Ridge | Medium | ||
| Linear regression likelihood function | DS AS MLE | Machine Learning | Linear regression, Likelihood function, Formula derivation | Medium | ||
| Model interpretability | DS AS MLE | Machine Learning | Interpretability | Easy | ||
| Multiclass evaluation metrics | DS AS MLE | Machine Learning | Multiclass, Diagnostics | Medium | ||
| NDCG vs. Mean Average Precision (MAP) vs. Mean Reciprocal Rank (MRR) | AS MLE | Machine Learning | Ranking metrics, NDCG, Mean average precision (MAP), Mean Reciprocal Rank (MRR), Recommender Systems | Medium | ||
| ROC vs. PR curve | AS MLE | Machine Learning | AUC, ROC, AUPR, Precision, Recall, Evaluation metrics | Medium | ||
| Measuring counterfactual impact | PDS DS | Metrics | Problem solving, Counterfactual | Hard | ||
| Sudden drop in user engagement | PDS DS | Metrics | Problem solving, Root-cause analysis | Medium | ||
| Documents edited by AI | PDS DS AS | Probability theory | Bayes rule, Conditional probability | Easy | ||
| Monty Hall | PDS DS AS | Probability theory | Bayes rule, Conditional independence, Prior evidence | Medium | ||
| Red and blue balls | PDS DS AS | Probability theory | Counting, Combinations, Repetition, Binomial | Easy | ||
| Trailing by two: should we go for two or three? | PDS DS AS | Probability theory | Independence, Decision making | Easy | ||
| Two children I | PDS DS AS | Probability theory | Prior evidence | Easy | ||
| Artist maxranks | PDS DS | SQL | Join | Easy | ||
| Histogram of songs | PDS DS | SQL | Recursive CTE, Join | Hard | ||
| Songs that did not enter the charts or entered high | PDS DS | SQL | Subquery, Join | Medium | ||
| Songs that stay in the chars for a while | PDS DS | SQL | Subquery, CTE, Join, ALL, Window functions | Medium | ||
| Gambler’s ruin win probability | PDS DS AS | Statistics | Gambler ruin, Random walk, Expectation | Medium | ||
| Manual estimation of flips | PDS DS AS | Statistics | Normal, CDF, Binomial, CLT | Medium | ||
| Measuring sticks | PDS DS AS | Statistics | Variance | Medium | ||
| Monotonic draws | PDS DS AS | Statistics | Expectation | Hard | ||
| Relationship between p-val and confidence interval | PDS DS AS | Statistics | Confidence interval, P-value, Hypothesis testing | Easy | ||
| Two-sample t-test | PDS DS AS | Statistics | Hypothesis testing | Easy | ||
| Choose a project | PDS DS AS MLE | Technical deep dive | Discussion prep | Hard | ||
| Questions worksheet (deep dive) | PDS DS AS MLE | Technical deep dive | Discussion prep | Hard | ||
| Interview success probability | PDS DS AS MLE | Application process | Application prep | Easy | ||
| Offer validity and cooling-off periods | PDS DS AS MLE | Application process | Application prep | Easy | ||
| Resume review | PDS DS AS MLE | Application process | Application prep, Resume | Easy | ||
| Up-level during interview | PDS DS AS MLE | Application process | Application prep | Easy | ||
| AUC ROC and predicted output transformations | DS AS MLE | Machine Learning | AUC ROC | Easy | ||
| Simulate dynamic coin flips | DS AS MLE | Machine Learning Coding | Simulation | Easy | ||
| Encoding categorical features | PDS DS AS MLE | Machine Learning | Categorical features, Embeddings, One-hot encoding, Hashing | Easy | ||
| Example of high-bias and high variance | DS AS MLE | Machine Learning | Bias-variance tradeoff | Medium | ||
| Range of R2 when combining regressions | PDS DS AS MLE | Machine Learning | Linear regression, Goodness of fit, R-squared, Correlation | Medium | ||
| Games between two players | PDS DS AS | Probability theory | Recursive relationship | Medium | ||
| Sum of normally distributed random variables | PDS DS AS | Probability theory | Normal, PDF, CDF | Medium | ||
| Unfair coin probability | PDS DS AS | Probability theory | Bayes rule, Conditional probability | Easy | ||
| Choose house or techno | PDS DS | SQL | Logical OR | Easy | ||
| Expensive house songs I | PDS DS | SQL | Subquery, CTE, Join | Medium | ||
| Expensive house songs II | PDS DS | SQL | Subquery, CTE, Join, Window functions | Hard | ||
| Songs that ranked 1 to 50 | PDS DS | SQL | Between | Easy | ||
| Biased coin | PDS DS AS | Statistics | Expectation, CLT, Binomial, Normal, Bernoulli, Hypothesis testing, CDF | Medium | ||
| Prussian horses | PDS DS AS | Statistics | Poisson, Hypothesis testing, CDF | Medium | ||
| AA tests | PDS DS AS | A/B Testing | Variance | Easy | ||
| Counterfactual definition | PDS DS AS | A/B Testing | Counterfactual | Easy | ||
| Equal-sized treatment and control groups | PDS DS AS | A/B Testing | Power, Variance, Sample size | Medium | ||
| False discovery control | PDS DS AS | A/B Testing | False discovery rate, Multiple hypotheses testing, Benjamini & Hochberg, Bonferroni | Easy | ||
| Interference | PDS DS AS | A/B Testing | Interference | Easy | ||
| Multi-armed and contextual bandits in AB testing | PDS DS AS | A/B Testing | Contextual bandits, Multi-armed bandits | Medium | ||
| Novelty and primacy effects | PDS DS AS | A/B Testing | Novelty effects, Primacy effects | Easy | ||
| Randomization checks | PDS DS AS | A/B Testing | Randomization | Easy | ||
| Randomization level | PDS DS AS | A/B Testing | Randomization, Variance | Medium | ||
| Activation functions | AS MLE | Machine Learning | Neural networks, Deep learning | Easy | ||
| Active learning | AS MLE | Machine Learning | Labels, Label sampling | Medium | ||
| Adagrad vs. RMSProp vs. Adam | AS MLE | Machine Learning | Neural networks, Deep learning, Optimization | Hard | ||
| Adaptive learning rate | AS MLE | Machine Learning | Neural networks, Deep learning, Optimization | Hard | ||
| Approximate nearest neighbors | AS MLE | Machine Learning | ANN, ANNOY, Nearest neighbors | Medium | ||
| Attention (intuition) | AS MLE | Machine Learning | Neural networks, Deep learning, Transformers, LLMs | Easy | ||
| Baselines | AS MLE DS | Machine Learning | Model evaluation | Easy | ||
| Bayesian Frequentist statistics | AS MLE | Machine Learning | Bayesian, Frequentist | Easy | ||
| Bias-variance biased estimator | DS AS MLE | Machine Learning | Bias-variance tradeoff | Medium | ||
| Bootstrap | PDS DS AS MLE | Machine Learning | Bootstrap | Easy | ||
| K-means from scratch | DS AS MLE | Machine Learning Coding | k-means | Medium | ||
| Linear regression with stochastic gradient descent | DS AS MLE | Machine Learning Coding | Stochastic Gradient descent, Linear regression | Medium | ||
| Logistic regression with gradient descent | DS AS MLE | Machine Learning Coding | Gradient descent, Logistic regression | Medium | ||
| Naive Bayes from scratch | DS AS MLE | Machine Learning Coding | Gaussian Naive Bayes | Medium | ||
| Neural network implementation | AS MLE | Machine Learning Coding | Gradient descent, Neural networks, Neuron | Hard | ||
| Principal Component Analysis (PCA) from scratch | DS AS MLE | Machine Learning Coding | Principal Component Analysis (PCA) | Medium | ||
| Common causes of data leakage | DS AS MLE | Machine Learning | Data leakage | Medium | ||
| Comparing decision trees with random forests | DS AS MLE | Machine Learning | Decision trees, Random forests | Easy | ||
| Correlation with binary variables | DS AS MLE | Machine Learning | Correlation, Hypothesis testing, Point-biserial correlation coefficient | Easy | ||
| Cross validation | PDS DS AS MLE | Machine Learning | Cross validation, Offline evaluation | Easy | ||
| Decide between a multinomial vs. a binary modeling approach | AS MLE | Machine Learning | Modeling, Multinomial, Binary | Easy | ||
| Discretization drawbacks | DS AS MLE | Machine Learning | Categorical variables, Discretization | Easy | ||
| Ensembles | AS MLE DS | Machine Learning | Ensembles, Numerical example | Medium | ||
| Examples of encoder and decoder models | AS MLE | Machine Learning | LLMs, Transformers, Encoder, Decoder | Easy | ||
| Exponentially weighted moving average | AS MLE | Machine Learning | Exponentially weighted moving average, Formula derivation, Proof | Medium | ||
| Feature crossing | AS MLE | Machine Learning | Feature engineering, Deep learning | Easy | ||
| Feature engineering in the era of deep learning | DS AS MLE | Machine Learning | Feature engineering | Easy | ||
| Gini impurity vs. information gain | AS MLE | Machine Learning | Decision tree, Information Gain, Gini impurity | Medium | ||
| Gradient boosting vs. random forests | DS AS MLE | Machine Learning | Gradient boosting, Random forests, Bagging, Boosting | Medium | ||
| Gradient descent vs. Stochastic Gradient descent and local minima | AS MLE | Machine Learning | Gradient descent (GD), Stochastic gradient descent (SGD), Local minima, Optimization, Deep learning | Medium | ||
| Gradient descent vs. Stochastic Gradient descent and learning rate | AS MLE | Machine Learning | Gradient descent (GD), Stochastic gradient descent (SGD), Learning rate, Optimization, Deep learning | Medium | ||
| How to get more labels | AS MLE | Machine Learning | Modeling, Label encoding | Medium | ||
| Hypothesis testing in regression coefficients | DS AS MLE | Machine Learning | Linear regression, Hypothesis testing | Medium | ||
| Information gain in decision trees | AS MLE | Machine Learning | Decision tree, Entroy, Information Gain, Formula derivation | Medium | ||
| Intercept | PDS DS AS MLE | Machine Learning | Linear regression, Intercept | Easy | ||
| Interpretability | PDS DS AS MLE | Machine Learning | ML interpretability | Easy | ||
| L2 regularization vs. weight decay | AS MLE | Machine Learning | Neural networks, Deep learning, Regularization | Hard | ||
| Linear regression assumptions | PDS DS AS MLE | Machine Learning | Linear regression | Easy | ||
| Linear regression with duplicated rows | DS AS MLE | Machine Learning | Linear regression, Statistical significance | Easy | ||
| Linear regression with stochastic gradient descent (formula derivation) | AS MLE | Machine Learning | Linear regression, Stochastic gradient descent, Formula derivation | Medium | ||
| Logistic regression and standardization | DS AS MLE | Machine Learning | Logistic regression, Standardization | Easy | ||
| Logistic regression assumptions | PDS DS AS MLE | Machine Learning | Logistic regression | Easy | ||
| Minimization of loss function intuition | AS MLE | Machine Learning | Neural networks, Deep learning, Optimization | Easy | ||
| Missing data | PDS DS AS MLE | Machine Learning | Missing data | Easy | ||
| Momentum | AS MLE | Machine Learning | Neural networks, Deep learning, Optimization | Hard | ||
| MSE vs. MAE | PDS DS AS MLE | Machine Learning | MSE, MAE | Easy | ||
| Multi-headed attention and self attention | AS MLE | Machine Learning | Neural networks, Deep learning, Transformers, LLMs | Medium | ||
| Multicollinearity | PDS DS AS MLE | Machine Learning | Multicollinearity, Linear regression | Medium | ||
| Negative sampling | AS MLE | Machine Learning | Neural networks, Deep learning, Negative sampling | Medium | ||
| Neuaral networks in layman terms | AS MLE | Machine Learning | Neural networks, Deep learning | Easy | ||
| Non-probability sampling | DS AS MLE | Machine Learning | Sampling, Non-probability | Easy | ||
| Normalization in neural networks | AS MLE | Machine Learning | Neural networks, Deep learning, Batch normalization, Layer normalization | Medium | ||
| Normalization vs. Standardization | PDS DS AS MLE | Machine Learning | Linear regression, Standardization, Normalization | Easy | ||
| Not enough data to train a model | DS AS MLE | Machine Learning | Data limitations | Easy | ||
| Optimize multiple objectives | AS MLE | Machine Learning | Modeling, Multiple objectives | Easy | ||
| Outliers | PDS DS AS MLE | Machine Learning | Outliers, Cook’s distance, Regularization | Easy | ||
| Overfitting in neural networks | AS MLE | Machine Learning | Neural networks, Deep learning, Overfitting | Medium | ||
| Positional embeddings | AS MLE | Machine Learning | Feature engineering, Deep learning, Transformers, LLMs, Positional embeddings, Positional encodings | Hard | ||
| Principal Component Analysis (PCA) | PDS DS AS MLE | Machine Learning | PCA | Easy | ||
| Prove that a median minizes MAE | AS MLE | Machine Learning | MAE, Median, Formula derivation, Proof | Hard | ||
| Random forest feature importance | AS MLE | Machine Learning | Feature importance, Explainability, Gini importance, Permutation importance | Medium | ||
| Random vs. stratified sampling | PDS DS AS MLE | Machine Learning | Sampling, Stratified sampling | Easy | ||
| Self-supervised learning | AS MLE | Machine Learning | Neural networks, Deep learning, Contrastive learning | Easy | ||
| SMOTE | AS MLE | Machine Learning | Imbalanced classification, SMOTE, Data augmentation | Easy | ||
| Transfer learning | AS MLE | Machine Learning | Neural networks, Deep learning, Transformers, LLMs, Catastrophic forgetting | Easy | ||
| Transfer learning vs. knowledge distillation | AS MLE | Machine Learning | LLM, Deep learning, Transfer learning, Knowledge distillation | Medium | ||
| Vanishing and exploding gradients (mathematical explaination) | AS MLE | Machine Learning | Neural networks, Deep learning, Mathematical explaination | Hard | ||
| Weight initialization | AS MLE | Machine Learning | Neural networks, Deep learning | Medium | ||
| Weighted and importance sampling | DS AS MLE | Machine Learning | Sampling, Weighted sampling, Importance sampling | Easy | ||
| Characteristics of metrics | PDS DS | Metrics | Characteristics of metrics | Easy | ||
| Types of metrics | PDS DS | Metrics | Types of metrics | Easy | ||
| Consecutive tails | PDS DS AS | Probability theory | Permutations, Repetition | Easy | ||
| Largest number rolled | PDS DS AS | Probability theory | Counting, Permutations, Repetition | Medium | ||
| Median probability | PDS DS AS | Probability theory | Binomial, Uniform, CDF | Medium | ||
| Number of emails | PDS DS AS | Probability theory | Poisson distribution | Easy | ||
| Paths to destination | PDS DS AS | Probability theory | Counting, Combinations | Easy | ||
| Repeated rolls until 4 | PDS DS AS | Probability theory | Geometric distribution | Easy | ||
| Sample digits 1-10 | PDS DS AS | Probability theory | Sample from samples, Uniform | Medium | ||
| Two fair die rolls | PDS DS AS | Probability theory | Independence, CDF, PMF | Easy | ||
| Artists with more songs than others | PDS DS | SQL | Subquery, CTE, Join, Window functions | Hard | ||
| Concat columns | PDS DS | SQL | Concat | Easy | ||
| Label recent songs | PDS DS | SQL | Case | Easy | ||
| Median songs per artist | PDS DS | SQL | CTE, Window functions | Hard | ||
| Songs in charts with greater durations | PDS DS | SQL | Subquery, CTE, Join, Window functions | Hard | ||
| Songs per genre | PDS DS | SQL | Group by | Easy | ||
| Songs with letters | PDS DS | SQL | Regexp | Easy | ||
| An intuitive way to write power | PDS DS AS | Statistics | Power, Hypothesis testing | Easy | ||
| Buy and sell stocks | PDS DS AS | Statistics | Gambler ruin, Expectation, Recursion, Random walk | Medium | ||
| CI of flipping heads | PDS DS AS | Statistics | Confidence Interval, CLT, Bernoulli trials | Medium | ||
| Confidence interval definition | PDS DS AS | Statistics | Confidence interval, Hypothesis testing | Easy | ||
| Confidence intervals that overlap | PDS DS AS | Statistics | Confidence interval, Hypothesis testing | Medium | ||
| Covariance of dependent variables | PDS DS AS | Statistics | Variance, Uniform, Covariance, Expectation | Medium | ||
| Distribution of a CDF | PDS DS AS | Statistics | CDF, Inverse transform | Medium | ||
| Dynamic coin flips | PDS DS AS | Statistics | Expectation, Simulation | Hard | ||
| Expected number of consecutive heads | PDS DS AS | Statistics | Expectation | Medium | ||
| Number of draws to get greater than 1 | PDS DS AS | Statistics | Normal, Geometric, CDF, Expectation | Medium | ||
| P-value definition | PDS DS AS | Statistics | P-value, Hypothesis testing | Easy | ||
| Tests for normality | PDS DS AS | Statistics | Hypothesis testing, Normality | Easy |