Welcome to the ultimate, comprehensive Data Science & AI Interview Preparation Repository! 🎯
This repository is designed as a one-stop hub for data science, machine learning, deep learning, MLOps, and Generative AI aspirants. It covers over 1,300+ curated interview questions and detailed answers across mathematics, machine learning theory, system design, coding, and production engineering.
🌐 Webpage Version: Ajit Singh - Data Science Interview Prep
All topics are organized into 4 core pillars. Click any module below to dive straight into the questions:
| Module | Focus Areas | Question Count |
|---|---|---|
| 🧮 Linear Algebra: Vectors & Matrices | Linear Algebra, Eigenvalues, SVD, Matrix Decompositions | 51 |
| 📚 Calculus & Algorithmic Differentiation | Gradients, Jacobians, Hessians, Automatic Differentiation | 41 |
| 🗂️ Information Theory | Entropy, Cross-Entropy, KL Divergence, Mutual Information | 31 |
| 📊 Descriptive Statistics | Summary Statistics, Variance, Skewness, Kurtosis | 52 |
| 🎲 Probability | Bayes Theorem, Random Variables, Distributions, Expectation | 106 |
| 🔍 Inferential Statistics | Hypothesis Testing, p-values, z/t-tests, Confidence Intervals | 58 |
| 🧪 A/B Testing & Experimentation | Power Analysis, MDE, CUPED, Network Effects, Peeking | 2 (New) |
| Module | Focus Areas | Question Count |
|---|---|---|
| 🧠 Machine Learning Fundamentals | Supervised/Unsupervised, Bias-Variance, ERM, Cross-Validation | 72 |
| 🛠️ Feature Engineering & Selection | Categorical Encoding, Scaling, Imputation, SHAP/LIME | 1 (New) |
| 📈 Linear Regression & Regularization | OLS, Ridge, Lasso, ElasticNet, Regression Diagnostics | 62 |
| 📉 Logistic Regression & Classification | Logistic Regression, Decision Boundaries, Odds Ratio, LDA | 32 |
| 🕸️ Support Vector Machines | Hard/Soft Margin, Kernel Trick, Dual Formulation | 39 |
| 🌲 Tree-Based Methods & Ensembles | Decision Trees, Random Forest, GBDT, XGBoost, LightGBM | 91 |
| 🔮 Unsupervised Learning | K-Means, Hierarchical, DBSCAN, PCA, t-SNE, UMAP | 98 |
| 🎲 Probabilistic Graphical Models | Naive Bayes, GMMs, EM Algorithm, Markov Chains | 49 |
| 🎯 Performance Metrics | ROC-AUC, PR-AUC, Precision/Recall, Log-Loss, MAE/RMSE | 118 |
| 📈 Time Series Analysis & Forecasting | ARIMA, Stationarity (ADF), Autocorrelation, Prophet | 150 |
| 🎮 Reinforcement Learning | MDPs, Q-Learning, DQN, Policy Gradients, Bandits | 1 (New) |
| Module | Focus Areas | Question Count |
|---|---|---|
| ⚡ Deep Learning Fundamentals | Activations, Backpropagation, Optimizers (Adam/SGD), Init | 134 |
| 🖼️ Convolutional Neural Networks | Convolutions, Pooling, ResNet, Object Detection, ViT | 46 |
| 🔄 Sequence Models & Transformers | RNNs, LSTMs, GRUs, Seq2Seq, Attention, Transformers | 9 |
| 🗣️ Natural Language Processing & LLMs | Tokenization, TF-IDF, Word2Vec, BERT, Language Models | 7 |
| 🤖 Generative AI & LLMs | LLM Architectures, LoRA/QLoRA, RAG, RLHF, DPO | 59 |
| Module | Focus Areas | Question Count |
|---|---|---|
| ⚙️ ML System Design | Two-Tower Models, Feature Stores, Real-time Serving, Drift | 7 |
| 📜 SQL for Data Science | Aggregations, Joins, Window Functions, CTEs, Subqueries | 47 |
| 🖥️ Computer Science & Coding | Algorithms, Big-O, Python Data Structures, Recursion, DP | 8 |
- 💡 Seamless GitHub Navigation: All questions feature standardized Markdown headers (
### Q: ...), generating direct clickable section anchors. - 🙈 Interactive Collapsible Answers: Answers are hidden inside
<details>tags so you can self-test before revealing the solution. - 📐 Clean Math Rendering: Formulas are formatted with native MathJax (
$and$$) for clean rendering on GitHub web and mobile. - 🚀 Zero Broken Links: All navigation links use robust relative repository paths.
Contributions are warmly welcome! If you'd like to add new questions, correct typos, or improve existing answers:
- Fork the Repository
- Create a Feature Branch (
git checkout -b feature/AddQuestions) - Commit your Changes (
git commit -m 'Add new LLM questions') - Push to the Branch (
git push origin feature/AddQuestions) - Open a Pull Request
Found an error or have suggestions? Reach out via:
- 📧 Email: sajit9285@gmail.com
- 🔗 LinkedIn: Ajit Singh
Happy Learning & Good Luck with your Data Science Interviews! 🎓🚀