Machine Learning Engineer Interview Questions

Showing 61–90 of 202 results

• InstacartHard·Machine Learning Engineer
  Explain Core ML Concepts

  This question tests the conceptual understanding of supervised learning foundations, the mathematical relationship between regression techniques, and the convergence behavior of optimization algorithms.

  Machine LearningTechnical ScreenLinear RegressionLogistic RegressionOptimization
• Jane StreetHard·Machine Learning Engineer
  Build a time-series forecasting model

  Build a time-series forecasting model 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 LearningTechnical Screen
• LinkedInMedium·Machine Learning Engineer
  Answer practical ML foundations questions

  This question evaluates an interviewee's practical mastery of ML foundations, covering probability calibration and its evaluation, feature selection methods including neural-network approaches, tokenizer types and trade-offs in NLP, and optimizer behavior such as Adam's maintained statistics.

  Machine LearningOnsite
• LinkedInMedium·Machine Learning Engineer
  Explain Core ML Fundamentals

  This question assesses core conceptual knowledge of supervised learning, optimization strategies, and regularization techniques used in model training.

  Machine LearningTechnical ScreenOptimizationRegularizationDecision Trees
• LinkedInMedium·Machine Learning Engineer
  Explain gradient descent, learning rate, backpropagation, overfitting and batch size

  A machine learning fundamentals screen on how neural networks are trained: how gradient descent updates parameters, what the learning rate and batch size change, how backpropagation computes gradients, and how to diagnose and fix underfitting and overfitting. Tests mechanics and practical training judgment, not definitions alone.

  Machine LearningTechnical Screen
• LinkedInEasy·Machine Learning Engineer
  Explain overfitting vs underfitting and fixes

  This question tests the conceptual understanding of the bias-variance tradeoff and the practical strategies used to improve model generalization.

  Machine LearningTechnical ScreenModel EvaluationBias-Variance TradeoffRegularization
• LinkedInHard·Machine Learning Engineer
  Implement a Text-Embedding Recommender Training Pipeline

  Tests the ability to structure an end-to-end, leakage-resistant training pipeline for a text-embedding recommender, covering profile flattening, embedding caching, temporal splits, binary classification, and artifact serialization.

  Machine LearningTechnical Screenrecommender-systemsfeature-engineeringnlp-embeddings
• LyftMedium·Machine Learning Engineer
  Explain Transformer and Fine-Tuning Basics

  This question evaluates understanding of transformer architectures, embedding models, and transfer learning concepts such as pretraining and fine-tuning, testing competence in representation learning and model adaptation.

  Machine LearningTechnical Screen
• LyftMedium·Machine Learning Engineer
  Implement K-Means Clustering from Scratch in Plain Python or NumPy

  A live coding exercise that asks you to implement k-means clustering from scratch in Python, with NumPy optional, and make it pass the interviewer's test cases. It tests the assignment and update steps, initialization, convergence, empty clusters and writing runnable code under time pressure.

  Machine LearningTechnical Screen
• LyftMedium·Machine Learning Engineer
  Reason About an Intercept Shift in Logistic Regression

  Tests understanding of how an additive shift to the intercept of a logistic regression model impacts predicted probabilities, ranking metrics (ROC-AUC), thresholded classification metrics (Recall, FPR, Precision), and calibration.

  Machine LearningTechnical ScreenLogistic RegressionModel EvaluationClassification Metrics
• LyftMedium·Machine Learning Engineer
  Train vs Test Error Diagnosis, Overfitting Fixes by Model Type, and Cross-Validation

  A machine learning fundamentals screen that asks you to diagnose models from pairs of training and test error, including the unusual case where the test error is the lower one. It continues with curbing overfitting in linear, tree-based and neural network models, and with setting up validation splits and k-fold cross-validation.

  Machine LearningTechnical Screen
• MercorMedium·Machine Learning Engineer
  Evaluate Noisy Data for LLM Post-Training

  Design a curation and validation framework to evaluate, filter, and safely integrate a large, noisy dataset into an LLM post-training pipeline without degrading model performance.

  Machine LearningTechnical ScreenLLM Post-TrainingData CurationModel Evaluation
• MercorMedium·Machine Learning Engineer
  Review a Compact GPT-Style Transformer Implementation

  Assess and debug a compact decoder-only Transformer implementation. Candidates trace tensor dimensions through causal multi-head self-attention, identify flaws in next-token loss alignment and KV-cached generation, and establish equivalence between cached and uncached forward passes.

  Machine LearningOnsiteTransformersPyTorchDeep Learning
• MetaMedium·Machine Learning Engineer
  Build harmful-content text classifier

  This problem evaluates your ability to design a robust text classification system for identifying harmful content, emphasizing strategies for data imbalance, labeling ambiguity, and real-time inference.

  Machine LearningOnsiteNLPClassificationMLOps
• MetaMedium·Machine Learning Engineer
  Derive Linear Regression Solution

  This problem tests your ability to derive the Ordinary Least Squares (OLS) closed-form solution for a simple linear regression model and analyze the underlying mathematical assumptions.

  Machine LearningTechnical ScreenLinear RegressionMathematicsMLE
• MetaHard·Machine Learning Engineer
  Implement 1NN Embeddings and Forward Pass

  Implement a fully vectorized forward pass that projects input features into an embedding space and classifies them using a 1-nearest-neighbor lookup against a reference bank.

  Machine LearningTechnical Screennumpydeep-learninglinear-algebra
• MetaMedium·Machine Learning Engineer
  Implement K-Means Clustering From Scratch

  An ML coding question that asks for a from-scratch implementation of k-means clustering with no provided test cases. It tests the assignment and update steps, vectorized distance computation, initialization, stopping rules, empty-cluster handling and the ability to verify code with self-constructed checks.

  Machine LearningTechnical Screen
• MetaMedium·Machine Learning Engineer
  Implement the Forward Pass of Multi-Head Attention From Scratch

  An ML coding question that asks for the forward pass of multi-head attention from scratch. It tests query, key and value projections, splitting and merging heads, scaled dot-product attention with a stable softmax, causal and padding masks, complexity in sequence length, and self-made correctness checks.

  Machine LearningTechnical Screen
• MetaHard·Machine Learning Engineer
  Machine Learning Fundamentals: Optimizers, Scaling Laws, and Clustering

  Tests deep conceptual mastery of modern ML core foundations: AdamW vs. Adam weight decay mechanics, compute-optimal Chinchilla scaling laws, and the mathematical equivalence between Hard Expectation-Maximization in GMMs and k-means clustering.

  Machine LearningOnsiteOptimizersScaling LawsClustering
• MetaMedium·Machine Learning Engineer
  Masked Multi-Head Attention Forward Pass Without a Framework

  Implement batched multi-head self-attention by hand, including projections, head split and merge, scaled dot products, a numerically stable masked softmax, and causal and padding masks, then prove it correct with tests you design yourself. Tests shape discipline and attention fundamentals.

  Machine LearningTechnical Screentransformersattentionnumerical-stability
• MetaHard·Machine Learning Engineer
  Self-Attention: Implementation, Complexity, and Efficient Variants

  Tests a candidate's mathematical and practical grasp of scaled dot-product attention, its computational and memory bottlenecks, and modern algorithmic optimizations like FlashAttention and linear attention.

  Machine LearningOnsitetransformerattention-mechanismflashattention
• MicrosoftMedium·Machine Learning Engineer
  Clean OCR data and build an LLM dataset

  This problem tests the ability to design an automated pipeline for cleaning noisy OCR text output and preparing it for supervised fine-tuning of an LLM.

  Machine LearningOnsiteData EngineeringOCRLLM Fine-tuning
• MicrosoftMedium·Machine Learning Engineer
  Compare preference alignment methods for LLMs

  This question tests your conceptual understanding of alignment techniques, your ability to evaluate trade-offs between stability and performance, and your approach to measuring model safety and utility.

  Machine LearningOnsiteRLHFDPOAlignment
• MicrosoftMedium·Machine Learning Engineer
  Explain metrics, regularization, and ablation studies

  This question evaluates understanding of classification evaluation metrics, regularization methods, and experimental design for component-level analysis in NLP and broader Machine Learning systems, testing competencies in performance measurement, regularization trade-offs, and causal attribution of model components.

  Machine LearningTechnical Screen
• MicrosoftMedium·Machine Learning Engineer
  Measure LLM Quality and Detect Hallucinations, Then Implement an Evaluation Component

  A hands-on LLM evaluation exercise on measuring model quality, comparing prompt or model versions, and detecting hallucinations against supplied context. It also asks for a small Python groundedness scorer that flags unsupported sentences and aggregates a hallucination rate, testing evaluation judgment and clean, well-tested code.

  Machine LearningOnsite
• Mistral AIMedium·Machine Learning Engineer
  Answer an LLM Systems Quick-Fire Interview

  This question tests your breadth of knowledge across the LLM stack, specifically focusing on inference optimization, evaluation strategies, and operational trade-offs.

  Machine LearningOnsiteLLMInferenceOptimization
• Mistral AIHard·Machine Learning Engineer
  Explain Transformer Internals and Implement Top-p Decoding

  Assess understanding of causal transformer decoding mechanics, including attention and KV caching, paired with an implementation of nucleus (top-p) sampling with temperature scaling.

  Machine LearningTechnical ScreenTransformersSampling AlgorithmsLLM Inference
• Mistral AIMedium·Machine Learning Engineer
  Optimize Nearest-Center Assignment Without Materializing NxKxD Distances

  This problem tests the ability to optimize memory usage in distance computations by leveraging algebraic expansion to avoid creating an intermediate (N, K, D) tensor.

  Machine LearningOnsiteNumPyVectorizationLinear Algebra
• NVIDIAMedium·Machine Learning Engineer
  Explain bias-variance, calibration, and model drift

  This question tests your conceptual understanding of model performance degradation over time, the significance of calibrated outputs for decision-making, and the fundamental trade-off between bias and variance.

  Machine LearningTechnical ScreenModel EvaluationProduction MLBias-Variance
• NetflixMedium·Machine Learning Engineer
  Compare Losses and Explain LoRA

  This question evaluates understanding of loss functions (mean squared error versus cross-entropy) and parameter-efficient fine-tuning via Low-Rank Adaptation (LoRA), emphasizing probabilistic interpretations, optimization behavior, and low-rank parameterization.

  Machine LearningTechnical Screen