Google interview questions

Showing 181–210 of 486 results

• Design multi-GPU matrix multiplicationHard

  This question evaluates proficiency in multi-GPU parallelism and system-level ML engineering, covering data partitioning, inter-GPU communication primitives, compute scheduling and overlap, memory layout and buffer reuse, numerical precision trade-offs, synchronization, scalability, and failure handling.

  GoogleMachine Learning EngineerML System DesignTechnical Screen
• Design nearest-available-taxi lookup with updatesMedium

  This question assesses the ability to design a high-concurrency geospatial system that maintains real-time state for moving entities and supports low-latency queries.

  GoogleSoftware EngineerSystem DesignTechnical Screen
• Design quota enforcement for high concurrencyHard

  This question evaluates a candidate's understanding of distributed system design and engineering competencies related to quota enforcement, high-throughput rate limiting, consistency models, sharding, and concurrency control under very high QPS.

  GoogleSoftware EngineerSystem DesignTechnical Screen
• Design relational-to-NoSQL migration pipelineHard

  This question evaluates a candidate's competency in designing a relational-to-NoSQL migration pipeline, covering distributed systems and data engineering concepts such as consistent snapshotting, change-data-capture, per-key ordering, idempotency, exactly-once behavior, schema evolution, partitioning, and operational reliability.

  GoogleSoftware EngineerSystem DesignTechnical Screen
• Design server allocation for multi-type nodesMedium

  This question evaluates understanding of scalable resource-allocation and matching systems, encompassing data-structure and index design, availability tracking, fragmentation minimization, concurrency control, and failure handling for large node pools.

  GoogleSoftware EngineerCoding & AlgorithmsTechnical Screen
• Design set with O(1) random accessMedium

  This question evaluates skill in designing a composite data structure that supports average O(1) insertion, deletion, and uniform random access, and tests understanding of hashing, indexing, randomness, and concurrency control.

  GoogleSoftware EngineerCoding & AlgorithmsTechnical Screen
• Design street-view image ingestion and storage systemHard

  This question tests the ability to design a high-throughput, fault-tolerant ingestion pipeline for massive geospatial media, focusing on distributed storage strategies, data integrity, and efficient retrieval at scale.

  GoogleSoftware EngineerSystem DesignTechnical Screen
• Design structure for insert and k-th largestMedium

  This problem tests your ability to design a dynamic data structure that maintains elements while efficiently querying the k-th largest value, necessitating a balance between insertion speed and retrieval performance.

  GoogleSoftware EngineerCoding & AlgorithmsOnsite
• Design tests to measure latency impactEasy

  A Google Data Scientist onsite case on measuring the causal impact of reduced YouTube video-start latency on engagement. It spans A/B design (primary metric, diagnostics, error/buffering guardrails), randomization unit and interference, power/MDE for heavy-tailed watch time, variance reduction (CUPED, stratification), analysis of conflicting movements, ratio-metric decomposition with Simpson’s paradox, and propensity score matching when randomization isn’t possible.

  GoogleData ScientistAnalytics & ExperimentationOnsite
• Design viewing history and resume serviceHard

  This question evaluates a candidate's ability in large-scale system design, covering distributed state management, API design, data modeling for per-user progress, idempotent update semantics, cross-device synchronization, and privacy/compliance considerations.

  GoogleSoftware EngineerSystem DesignTechnical Screen
• Detect and Prevent Reward HackingHard

  Assess the candidate's understanding of reward hacking (specification gaming) in reinforcement learning from human feedback (RLHF), focusing on diagnosing exploitation patterns, designing detection metrics, applying conservative policy updates, and implementing recovery safeguards.

  GoogleMachine Learning EngineerMachine LearningTechnical Screen
• Determine All Players with Fixed RankingsMedium

  Determine which players in a competition have a uniquely determined ranking given a set of pairwise match outcomes.

  GoogleSoftware EngineerCoding & AlgorithmsOnsite
• Determine Impact of New Chat-Notification on User EngagementMedium

  Determine Impact of New Chat-Notification on User Engagement evaluates statistical assumptions, formulas, estimation strategy, uncertainty, edge cases, and interpretation in a realistic interview setting. A strong answer states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

  GoogleData ScientistStatistics & MathTechnical Screen
• Determine Normality of Single Observation with Z-TestEasy

  Evaluates hypothesis testing for a single observation against a fully specified standard normal distribution. Strong answers state hypotheses, use z = x, define critical values and p-values, and note limitations.

  GoogleData ScientistStatistics & MathTechnical Screen
• Determine Reporting RelationshipsMedium

  This question evaluates understanding of hierarchical relationship modeling, transitive inference across relations, and dynamic data-structure design for tracking manager and peer links.

  GoogleSoftware EngineerCoding & AlgorithmsTechnical Screen
• Determine Whether a Word Exists in a GraphMedium

  This problem tests the ability to implement a depth-first search (DFS) with backtracking to find a specific sequence within a grid, while managing state to avoid reusing the same path.

  GoogleSoftware EngineerCoding & AlgorithmsOnsite
• Determine validity after digit-constrained deletionsMedium

  This interview question evaluates algorithm design, data structures, correctness, complexity, edge cases, and implementation details in a realistic interview setting. A strong answer for Determine validity after digit-constrained deletions states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

  GoogleSoftware EngineerCoding & AlgorithmsOnsite
• Determine whether two nodes are relatedMedium

  This problem tests your ability to determine connectivity between two nodes in a directed graph, specifically modeling ancestry relationships.

  GoogleSoftware EngineerCoding & AlgorithmsTechnical Screen
• Determine winner in stack-merging gameHard

  This problem tests your ability to model a combinatorial game using the Sprague-Grundy theorem and memoization to determine the winner of a state-dependent stack-merging process.

  GoogleSoftware EngineerCoding & AlgorithmsTechnical Screen
• Deterministic Task Execution OrderEasy

  This question tests the ability to implement Kahn's algorithm for topological sorting while using a min-priority queue to satisfy specific tie-breaking requirements.

  GoogleSoftware EngineerCoding & AlgorithmsOnsite
• Diagnose Google Meet Disconnections and Assess Business ImpactHard

  Evaluates reliability analytics for Google Meet disconnections and enterprise renewal impact. Strong answers define call drops, use client and server logs to localize root causes, estimate account-level retention impact, and prioritize fixes using quantified business risk.

  GoogleData ScientistAnalytics & ExperimentationTechnical Screen
• Diagnose a metric drop in search timeMedium

  This question evaluates a data scientist's competency in precise metric and population definition, time‑to‑event and survival/hazard modeling, time‑series decomposition and change‑point analysis, instrumentation and traffic forensics, and experimental design for root‑cause identification.

  GoogleData ScientistAnalytics & ExperimentationOnsite
• Diagnose and fix flawed model fitHard

  This question tests your ability to diagnose poor model performance in a binary classification task involving imbalanced data, high-cardinality categorical features, and potential bias.

  GoogleData ScientistMachine LearningTechnical Screen
• Diagnose and reverse an adoption-rate declineHard

  This question evaluates a data scientist's competency in product analytics and experimentation, covering precise metric definition and validation, instrumentation and event quality checks, rapid triage of adoption regressions, hypothesis generation, experiment design, and causal inference.

  GoogleData ScientistAnalytics & ExperimentationOnsite
• Discuss Challenges, Difficult Colleagues, and MultitaskingMedium

  This question assesses your ability to navigate workplace friction and manage competing priorities through structured reflection and professional communication.

  GoogleSoftware EngineerBehavioral & LeadershipOnsite
• Discuss Complex Systems and Failure ExamplesMedium

  This question assesses a candidate's ability to analyze distributed system failures, communicate technical post-mortems clearly, and demonstrate personal accountability and iterative improvement.

  GoogleSoftware EngineerBehavioral & LeadershipTechnical Screen
• Earliest Time of Full ConnectivityMedium

  Determine the earliest timestamp at which all nodes in a network become part of a single connected component using a chronological sequence of edge additions.

  GoogleSoftware EngineerCoding & AlgorithmsOnsite
• Enforce a Country's IP Ban List on Incoming Connections Using Its Lookup APIMedium

  A system design exercise where a country provides an API reporting whether an IP address is banned and the service must refuse connections from banned addresses. It tests scoping questions such as IPv6 and address ranges, fast local enforcement at the edge, keeping ban data fresh, and fail-safe compliance behavior.

  GoogleSoftware EngineerSystem DesignOnsite
• Estimate Population Mean and Conversion Rate AccuratelyHard

  Evaluates core statistical inference skills across p-values, confidence intervals, standard error reduction, tail probability estimation, efficient conversion-rate sampling, and truncated-normal estimation. Strong answers state assumptions, use appropriate intervals and weights, and avoid common interpretation errors.

  GoogleData ScientistStatistics & MathOnsite
• Estimate percentile from bucketsMedium

  Estimate a percentile from histogram buckets by locating the cumulative-count bucket, explaining why midpoint can be poor, interpolating within the bucket under explicit assumptions, handling log-spaced buckets and edge cases, and reporting uncertainty when buckets are coarse.

  GoogleData ScientistStatistics & MathTechnical Screen