practice a little every day, paws crossed ♡
Data Scientist Interview Questions
Practice 9403 coding, system design and machine learning questions from 546 companies, sorted by role, topic and interview round. Every question has step-by-step hints, a code runner that supports 8 languages and an AI mentor; members also get the full solution walkthrough.
- 9403
- Questions
- 546
- Companies
- 11
- Topics
- 8
- Languages
Showing 271–300 of 1971 results
- ByteDanceMediumData ScientistCount Same-Day Registered Posters
This question evaluates SQL querying skills, including joining relational tables, date handling, aggregation, and deduplicating users across events. Commonly asked in Coding & Algorithms interviews for data scientists, it assesses practical relational database querying and analytics tasks—filtering, grouping, and counting distinct users across tables—and tests practical application of SQL rather than purely conceptual database theory; this summary is in English.
Coding & AlgorithmsTechnical Screen - ByteDanceMediumData ScientistDetermine Whether an Integer Is a Palindrome
This problem tests your ability to manipulate integers mathematically to detect palindromic sequences without converting the number into a string, requiring careful handling of edge cases and integer overflow.
Coding & AlgorithmsTechnical ScreenMathAlgorithmsInteger Manipulation - ByteDanceMediumData ScientistDiagnose Low CTR in an Advertising Campaign Funnel
Prepare for a product analytics interview by diagnosing below-average advertising CTR from limited campaign-level funnel metrics. The question tests metric interpretation, uncertainty, evidence-versus-hypothesis discipline, and how to plan useful follow-up analysis.
Analytics & ExperimentationTechnical Screen - ByteDanceMediumData ScientistDiscuss Your Data Analytics and Causal-Inference Experience
Practice describing data analytics and causal-inference experience while separating descriptive insights from credible intervention effects.
Behavioral & LeadershipTechnical Screen - ByteDanceMediumData ScientistExplain Type I/II errors vs precision/recall
This question evaluates understanding of Type I and Type II errors in hypothesis testing and binary classification, their correspondence to false positives and false negatives, and how those concepts relate to precision and recall within the Statistics & Math domain.
Statistics & MathTechnical Screen - ByteDanceMediumData ScientistFind Each User's Earliest Seven-Day Activity Streak
Tests your ability to identify consecutive date sequences (gaps-and-islands problem) in PostgreSQL, aggregate streaks, and filter for each user's earliest qualifying period.
Data Manipulation (SQL/Python)Technical Screensqlpostgresqlwindow-functions - ByteDanceMediumData ScientistFind top-paid employee per department
This problem tests your ability to perform group-wise filtering and ranking using window functions or joins to extract top-performing records within categories.
Data Manipulation (SQL/Python)Technical ScreenSQLWindow FunctionsAggregation - ByteDanceHardData ScientistHow 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 - ByteDanceMediumData ScientistHow would you choose a classification threshold?
This question evaluates understanding of threshold selection for probabilistic binary classifiers, testing competencies in validation-set decision threshold choice, cost-sensitive trade-offs between false positives and false negatives, effects of class imbalance, and probability calibration.
Machine LearningTechnical Screen - ByteDanceEasyData ScientistImplement several OA simulation problems
This set of problems evaluates skills in string processing, memory simulation and allocation logic, and online data-structure maintenance for dynamic contiguous segment counting, testing algorithmic design and attention to stateful invariants.
Coding & AlgorithmsOnline Assessment - ByteDanceHardData ScientistImplement stack variants and upward path sum
A four-part ByteDance Data Scientist onsite coding question: implement an O(1) MinStack, a MaxStack with peekMax/popMax, a streaming median finder, and a binary-tree upward (node-to-ancestor) path-sum check. It tests data-structure augmentation, streaming order statistics, and prefix-sum tree reasoning.
Coding & AlgorithmsOnsite - ByteDanceHardData ScientistInterpret and validate regression with interactions
This question evaluates understanding of regression modeling with interaction terms, binary outcome models and odds/marginal-effect interpretation, causal inference concepts like endogeneity and instrumental variables, model diagnostics and standard-error choices, and count-data modeling.
Statistics & MathTechnical Screen - ByteDanceHardData ScientistPredict Short-Video Engagement with Skewed Data
Design a short-video engagement model with explicit targets, leakage-safe features, heavy-tail handling, exposure awareness, and temporal evaluation.
Machine LearningTechnical Screen - ByteDanceMediumData ScientistProbability That One Fair Die Beats Another
Solve a coding interview problem about the probability that one fair die beats another when the dice may have different sizes and duplicate face values. The challenge emphasizes strict tie handling and an efficient approach that avoids comparing every possible pair of faces.
Coding & AlgorithmsTechnical Screen - C3 AIMediumData ScientistChoose Tokenization and Dense or Sparse Retrieval
This problem tests your understanding of document retrieval strategies, specifically comparing sparse and dense representations and how to integrate them into a production RAG system.
Machine LearningTechnical ScreenRAGNLPInformation Retrieval - C3 AIMediumData ScientistChoose a Parameter to Reduce Underfitting in Gradient-Boosted Trees
Explain why decreasing minimum observations per leaf can reduce gradient-boosted tree underfitting and how to validate the change.
Machine LearningOnline Assessment - C3 AIMediumData ScientistCompare How Models Respond to Multicollinearity
Compare multicollinearity effects in regression, Lasso, logistic regression, decision trees, and boosted trees without overstating immunity.
Machine LearningOnline Assessment - C3 AIHardData ScientistCompare PCA and Autoencoders Under the Curse of Dimensionality
This question tests your conceptual understanding of dimensionality reduction techniques and your ability to weigh their performance in high-dimensional feature spaces prone to the curse of dimensionality.
Machine LearningTechnical ScreenDimensionality ReductionPCAAutoencoders - C3 AIHardData ScientistCompare Transformers and RNNs and Derive Attention Complexity
Evaluate candidate's deep learning foundations by comparing Transformer and recurrent architectures and deriving exact time/memory complexities for standard scaled dot-product attention across training and inference.
Machine LearningTechnical ScreenTransformersRNNAttention Mechanism - C3 AIHardData ScientistConnect Bias-Variance Trade-offs to Bagging and Boosting
This question tests the conceptual understanding of how ensemble methods manipulate model error components through different structural paradigms.
Machine LearningTechnical ScreenBias-VarianceEnsemble LearningBagging - C3 AIMediumData ScientistDecompress an encoded string
Decompress an encoded string evaluates algorithm design, data structures, correctness, complexity, edge cases, and implementation details in a realistic interview setting. A strong answer states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.
Coding & AlgorithmsOnsite - C3 AIMediumData ScientistDescribe challenge, influence, and mentorship outcomes
This question evaluates leadership competencies for a Data Scientist role, including influence and persuasion, mentoring, prioritization, trade-off reasoning, and the ability to quantify impact through metrics.
Behavioral & LeadershipOnsite - C3 AIHardData ScientistEvaluate a Retrieval-Augmented Generation System
Design an end-to-end evaluation framework for an enterprise RAG system that assesses retrieval quality, generative fidelity, claim-level attribution, security controls, and production telemetry.
ML System DesignTechnical ScreenRAGLLM EvaluationSystem Design - C3 AIMediumData ScientistExplain Training R-Squared After Adding a Random Feature
Prove that ordinary training R-squared cannot decrease when adding a feature, while adjusted and test-set R-squared can.
Statistics & MathOnline Assessment - C3 AIHardData ScientistExplain an End-to-End ML Project to Mixed Audiences
This question assesses the ability to articulate a technical machine learning project, balancing depth for engineers with accessibility for non-technical stakeholders, while demonstrating rigorous problem-solving.
Behavioral & LeadershipOnsiteBehavioralML LifecycleCommunication - C3 AIMediumData ScientistExplain 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 - C3 AIMediumData ScientistExplain the Gaussian Distribution and Its Limits
Tests a candidate's mathematical understanding of the Gaussian distribution, key theoretical properties such as affine closure and the Central Limit Theorem, and the ability to diagnose and handle non-normal distributions in real-world modeling.
Statistics & MathTechnical ScreenStatisticsProbabilityAnomaly Detection - C3 AIMediumData ScientistExplain the Margin View of Boosting Classification
Understand boosting classification through sequential learners, signed margins, surrogate losses, and the limits of training-margin claims.
Machine LearningOnline Assessment - C3 AIHardData ScientistFind longest substring with at least k repeats
This interview question evaluates algorithm design, data structures, correctness, complexity, edge cases, and implementation details in a realistic interview setting. A strong answer for Find longest substring with at least k repeats states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.
Coding & AlgorithmsTechnical Screen - C3 AIHardData ScientistFind the Intersection of Two Linked Chains
This problem tests the ability to detect the intersection point of two singly linked lists with constant space complexity, focusing on pointer manipulation and list traversal techniques.
Coding & AlgorithmsOnsiteLinked ListsTwo PointersAlgorithms
C3 AI