AI & machine learning interview prep
Everything for data scientist, ML engineer and AI engineer interviews, in the order that makes it click: 5 chapters, 250 questions with model answers, cheat sheets and in-depth guides. Then practise out loud with the voice coach.
Statistics & Data Science
Probability, distributions and hypothesis tests: the maths every later step leans on, and its own interview round for data roles.
- A/B Test Sample Size and the Peeking Problem: Interview Guide
- Bayes' Theorem and the Base Rate Fallacy: Interview Guide
- Central Limit Theorem Explained: Data Science Interview Guide
- Confidence Intervals Explained: Data Science Interview Guide
- Linear Regression Assumptions: What Breaks and How to Check
- P-Values and Hypothesis Testing: Data Science Interview Guide
Machine Learning
How models learn and how to tell when they generalize: bias-variance, metrics, validation and the classic algorithms.
- Bias-Variance Tradeoff Explained: ML Interview Guide
- Cross-Validation and Data Leakage in Machine Learning
- Gradient Descent Explained: Batch vs SGD vs Mini-Batch
- Imbalanced Classification: Class Weights, SMOTE and PR-AUC
- Logistic Regression Interview Questions: Sigmoid, Log Loss, Odds
- Overfitting and L1 vs L2 Regularization Explained with Code
Deep Learning
Neural networks from backpropagation to transformers: the machinery inside every modern model.
- SGD, Momentum, Adam and AdamW: Optimizer Interview Guide
- Backpropagation Explained: The Chain Rule, Step by Step
- Batch Norm vs Layer Norm: Deep Learning Interview Question
- CNN Interview Guide: Convolution, Pooling and Receptive Field
- Dropout and Overfitting in Deep Neural Networks Explained
- RNN vs LSTM vs GRU: How Gated Recurrent Networks Work
AI & LLM Engineering
Building products on large language models: prompting, embeddings, RAG, agents and evaluation.
- Embeddings and Vector Search: AI Engineer Interview Guide
- How to Evaluate LLM Applications: Evals and LLM Judges
- Fine-Tuning vs RAG: When to Use Each in LLM Apps
- LLM Function Calling and Agent Loops Explained
- Prompt Injection Attacks and Defenses for LLM Apps
- How to Design a RAG Pipeline: Chunking, Retrieval and Prompts
MLOps
Getting models into production and keeping them healthy: pipelines, serving, monitoring and drift.
- Batch vs Online Inference: Choosing How to Serve a Model
- CI/CD for Machine Learning: Pipelines and Continuous Training
- Data Drift vs Concept Drift: Detecting Drift with PSI and KS
- LLMOps: Prompt Versioning and Evaluation Gates in CI
- MLflow Experiment Tracking and Model Registry Explained
- Reproducible ML: Versioning Data, Code and Models