Machine Learning Interview Questions

Showing 31–60 of 163 results

• AnthropicMedium·Machine Learning Engineer
  Explain batch inference design

  This question tests your ability to design a robust, production-grade batch inference pipeline, focusing on data consistency, system reliability, and scalability.

  Machine LearningOnsiteSystem DesignMLOpsBatch Processing
• AnthropicMedium·Machine Learning Engineer
  Fill In the Sampling and Decode Loop of an LLM Inference Notebook

  Anthropic Fellows final-round practice: complete temperature/top-k/top-p sampling and a batched KV-cache decode loop in a Colab notebook so greedy output matches model.generate.

  Machine LearningOnsiteAnthropic FellowsAI Safety Fellowshipllm-inference
• AnthropicHard·Software Engineer
  Implement and analyze custom attention

  This question evaluates implementation and analysis skills for scaled dot-product attention, testing competencies in efficient tensorized PyTorch coding, numerical stability and mixed-precision handling, masking semantics (causal and padding), multi-head shape correctness, and unit-test validation including gradient and edge-case behavior.

  Machine LearningOnsite
• AnthropicMedium·Software Engineer
  Implement and derive backprop from scratch

  This question evaluates understanding and practical implementation of neural network fundamentals, specifically analytic backpropagation and gradient derivation, numerically stable binary cross-entropy computation, parameter initialization, gradient descent updates, and gradient checking.

  Machine LearningOnsite
• AppleMedium·Machine Learning Engineer
  Analyze vision model failures

  This question evaluates proficiency in computer vision machine learning fundamentals (including bias–variance tradeoffs, overfitting, class imbalance, evaluation metrics, threshold selection, calibration, and data leakage), probabilistic reasoning for performance interpretation, and production model diagnosis and root-cause analysis.

  Machine LearningOnsite
• AppleMedium·Data Scientist
  Explain CNN shapes, params, and trade-offs

  This question tests your ability to calculate output dimensions, parameter counts, and computational costs (FLOPs/MACs) in CNNs, while requiring you to justify architectural choices like depthwise convolutions and normalization layers.

  Machine LearningOnsiteCNNDeep LearningModel Architecture
• AppleMedium·Machine Learning Engineer
  Explain classification lifecycle and CTR modeling

  This question evaluates competency in end-to-end supervised classification and production machine learning systems, covering text feature engineering (bag-of-words), ensemble evaluation via out-of-bag methods, CTR prediction workflows including calibration and evaluation, and operational debugging of deployed models.

  Machine LearningOnsite
• AppleHard·Data Scientist
  Implement random forest with OOB and imbalance

  This question evaluates expertise in implementing and engineering a memory- and compute-efficient Random Forest for binary classification, covering ensemble methods, CART/Gini impurity, OOB evaluation and calibration, class-imbalance handling, reproducibility, parallel and thread-safe system design, and streaming/warm-start model updates.

  Machine LearningOnsite
• BByteDanceMedium·Software Engineer
  Hand-Write a Multi-Head Attention Layer: Projections, Head Splitting and Masking

  Write a multi-head attention layer from scratch: project queries, keys and values, split them into heads, apply scaled dot-product attention with optional padding or causal masks, and merge the heads. It tests tensor shape discipline, numerical stability, and the cost of attention as sequence length grows.

  Machine LearningOnsite
• BByteDanceHard·Data Scientist
  How to deploy and tune multimodal models?

  A six-part ByteDance data scientist onsite ML interview covering deploying and tuning multimodal models under GPU/VRAM, compute, and latency constraints; designing fast video retrieval over precomputed captions and embeddings with ANN and hybrid reranking; and deep-learning fundamentals — overfitting detection and mitigation, dropout and inverted-dropout scaling, normalization layers (BatchNorm/LayerNorm/GroupNorm/InstanceNorm/RMSNorm), and reinforcement learning for LLM post-training (RLHF). It tests practical systems tradeoffs alongside theoretical foundations across ML systems, retrieval, and reinforcement learning.

  Machine LearningOnsite
• C3 AIMedium·Data Scientist
  Explain key ML/stats concepts

  Explain key ML/stats concepts evaluates core ML concepts, assumptions, math intuition, training/evaluation trade-offs, and practical failure modes in a realistic interview setting. A strong answer states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

  Machine LearningOnsite
• Capital OneEasy·Data Scientist
  Build House Price Model Responsibly

  This problem tests your ability to design a robust, fair, and scalable machine learning pipeline for house price prediction, emphasizing data preprocessing, model selection, and ethical oversight.

  Machine LearningOnsiteSupervised LearningRegressionFairness
• Capital OneMedium·Data Scientist
  Build and evaluate donation propensity model

  This question evaluates a data scientist's competency in profit-oriented predictive modeling, including conversion and donation-amount modeling, handling sparse and categorical features, calibration and uplift estimation, and offline policy evaluation under selection bias.

  Machine LearningOnsite
• Capital OneMedium·Data Scientist
  Design ML deployment with GitHub and Jenkins

  This question tests the ability to design an end-to-end MLOps pipeline, focusing on CI/CD orchestration, model versioning, reproducibility, and security compliance in a regulated financial environment.

  Machine LearningOnsiteMLOpsCI/CDSystem Design
• Capital OneHard·Data Scientist
  Design a production face recognition system

  This question evaluates production ML system design and engineering skills specific to face recognition, covering on-device model selection and embedding design, training objectives and evaluation protocols, anti-spoofing and robustness mechanisms, privacy and template protection, performance/resource constraints, fairness and monitoring, and safe experimentation. It is asked in the Machine Learning domain for Data Scientist roles because it assesses trade-offs between privacy, security, latency, and scalability in real-world systems, and it tests both conceptual understanding of biometric and security principles and practical application for deploying and operating on-device ML.

  Machine LearningOnsite
• Capital OneMedium·Data Scientist
  Diagnose Multicollinearity in Flight Delay Prediction Model

  Evaluates flight-delay modeling, data quality checks, and multicollinearity diagnosis. Strong answers identify leakage and timestamp issues, choose classification or regression based on the target, diagnose VIF and correlated features, and mitigate multicollinearity based on prediction or interpretation goals.

  Machine LearningOnsite
• Capital OneMedium·Data Scientist
  Evaluate Models for Credit-Risk Scoring at Capital One

  This interview question evaluates core ML concepts, assumptions, math intuition, training/evaluation trade-offs, and practical failure modes in a realistic interview setting. A strong answer for Evaluate Models for Credit-Risk Scoring at Capital One states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

  Machine LearningOnsite
• Capital OneMedium·Data Scientist
  Evaluate OutlierHandler Class for Code Quality and Testing

  Evaluates code-review judgment for OutlierHandler and imputer preprocessing classes in ML pipelines. Strong answers explain fit-transform separation, leakage prevention, API quality, edge cases, and testing.

  Machine LearningOnsite
• Capital OneMedium·Data Scientist
  Evaluate Python Class Design in Data Pipeline

  This interview question evaluates core ML concepts, assumptions, math intuition, training/evaluation trade-offs, and practical failure modes in a realistic interview setting. A strong answer for Evaluate Python Class Design in Data Pipeline states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

  Machine LearningOnsite
• Capital OneMedium·Machine Learning Engineer
  Explain core ML concepts and lifecycle

  This question evaluates understanding of core machine learning concepts and lifecycle competencies, including bias–variance tradeoff, parameters versus hyperparameters, batch versus real-time inference, strategies for training at scale, feature engineering, and end-to-end model development.

  Machine LearningOnsite
• Capital OneMedium·Data Scientist
  Identify Risks and Improve Imputation Class Implementations

  This interview question evaluates core ML concepts, assumptions, math intuition, training/evaluation trade-offs, and practical failure modes in a realistic interview setting. A strong answer for Identify Risks and Improve Imputation Class Implementations states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

  Machine LearningOnsite
• CohereMedium·Machine Learning Engineer
  Defend Your Own Research Paper: Experimental Setup, Practicality and Later Work

  A research paper round where you present your own published paper to an interviewer who already knows it and questions its weaknesses. Prepare to judge whether the experimental setup was reasonable, whether the results are practical, and what the field has learned since publication, all with specific, honest self-critique.

  Machine LearningOnsite
• CohereMedium·Machine Learning Engineer
  Explain LLM training, RL, and evaluation

  This question tests your conceptual understanding of the end-to-end LLM lifecycle, specifically the transition from raw text pre-training to alignment and evaluation.

  Machine LearningOnsiteLLMRLHFPPO
• CohereMedium·Machine Learning Engineer
  Explain Top-k and Top-p Decoding

  This question tests your conceptual and mathematical understanding of Top-k and Top-p (nucleus) sampling, focusing on how they modify the output probability distribution of autoregressive language models to balance creativity and coherence.

  Machine LearningOnsiteNLPGenerative AISampling
• CoinbaseMedium·Machine Learning Engineer
  Build a baseline classification model from messy data

  This question evaluates skills in practical machine learning engineering, including data cleaning, preprocessing, feature selection, handling mixed numeric and categorical features, and baseline model construction for a binary classification task.

  Machine LearningOnsite
• DatabricksMedium·Machine Learning Engineer
  Implement Gradient Descent Regression

  This problem tests your understanding of linear regression by requiring the implementation of a gradient descent optimizer from scratch to minimize the Mean Squared Error loss.

  Machine LearningOnsiteLinear RegressionGradient DescentOptimization
• DatabricksMedium·Machine Learning Engineer
  Implement Linear Regression Gradient Descent

  This question tests the ability to implement batch gradient descent for simple linear regression from scratch, requiring the derivation of gradients for weights and bias, and understanding how to optimize convergence.

  Machine LearningOnsiteGradient DescentLinear RegressionOptimization
• DatabricksMedium·Machine Learning Engineer
  Implement Simple Linear Regression with Gradient Descent from Scratch

  Write simple linear regression trained with gradient descent from scratch, choosing the loss, learning rate and stopping rule yourself. Tests deriving the gradients correctly, writing a runnable training loop, understanding feature scaling and convergence, and verifying results against a closed-form fit.

  Machine LearningOnsite
• DoorDashHard·Data Scientist
  Build ETA prediction and simulate impact

  This question evaluates machine learning competencies for ETA prediction, including feature engineering, leakage detection, time-based validation, point and quantile modeling, evaluation and calibration metrics, cost-sensitive decisioning, explainability and fairness auditing, and production deployment considerations.

  Machine LearningOnsite
• DoorDashHard·Data Scientist
  Build a late-delivery risk model

  This question evaluates competencies in production-ready predictive modeling, including target definition, feature engineering, temporal train/validation splitting to avoid leakage, model evaluation and calibration, monitoring and ramp strategies, and translating probabilistic outputs into business-impacting decisions for a Data Scientist.

  Machine LearningOnsite