practice a little every day, paws crossed ♡
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 2881–2910 of 4616 results
- OneMain FinancialEasyData ScientistExplain decision trees and tree ensembles
This question evaluates understanding of decision trees, splitting criteria for classification versus regression, hyperparameter effects on the bias–variance trade-off, and the mechanics and rationale behind tree-based ensemble methods in the Machine Learning domain.
Machine LearningTechnical Screen - OneMain FinancialMediumData ScientistMaximize Probability of Drawing Two Red Balls
This interview question evaluates statistical assumptions, formulas, estimation strategy, uncertainty, edge cases, and interpretation in a realistic interview setting. A strong answer for Maximize Probability of Drawing Two Red Balls states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.
Statistics & MathTechnical Screen - OneMain FinancialMediumData ScientistMaximize credit card portfolio profit
This problem tests your ability to model expected profitability by balancing credit risk (probability of default) against revenue generation and defining optimal decision thresholds.
Analytics & ExperimentationTechnical ScreenCredit RiskProfit ModelingDecision Strategy - OneMain FinancialMediumData ScientistOptimize SaaS pricing and profit
This question evaluates pricing optimization, linear demand modeling, capacity-constrained profit maximization, and the impact of subscription churn on lifetime value within an Analytics & Experimentation data-science context.
Analytics & ExperimentationTechnical Screen - OneMain FinancialHardData ScientistPresent a project to non-technical leaders
This question evaluates a data scientist's ability to communicate a modeling project's business framing, metric choices, trade-offs, risk mitigation, and quantified impact while handling cross-functional questions.
Behavioral & LeadershipTechnical Screen - OneMain FinancialHardData ScientistSelect and tune XGBoost hyperparameters
This question evaluates skills in selecting and tuning XGBoost hyperparameters, managing severe class imbalance and sparse one‑hot encodings, handling missing values, and designing compute‑efficient training and grouped cross‑validation to prevent user‑level leakage.
Machine LearningTechnical Screen - OneMain FinancialEasyData ScientistWalk through a DS project end-to-end
This question evaluates end-to-end data science competencies including problem framing, metric design, data sourcing and quality assessment, modeling and evaluation choices, risk identification, impact quantification, and stakeholder communication within the Behavioral & Leadership category for a Data Scientist position.
Behavioral & LeadershipTechnical Screen - OnepayEasySoftware EngineerCount Complete Highway Journeys from Toll Logs
This problem tests string manipulation, hash map usage, and state tracking to identify valid entry-to-exit sequences from an unordered stream of toll logs.
Coding & AlgorithmsTechnical ScreenHash MapsSortingData Streams - OnepayMediumSoftware EngineerDebug and Extend a Device Registry: Status Validation, Active Counts, Bulk Updates
A progressive coding exercise on a device registry: find the bug behind a failing test where a device's status must be allowed for its type, then add validated registration, per-type counts of active devices that omit zeros, and a bulk status update based on maintenance time and failure count. It tests code reading and business logic.
Software Engineering FundamentalsTechnical Screen - OnepayEasySoftware EngineerDesign a Discord-Style Communication Platform
This problem tests your ability to design a scalable real-time messaging architecture, focusing on handling message persistence, state synchronization for offline clients, and channel-based message distribution.
System DesignTechnical ScreenReal-timeScalabilityMessaging - OpenAIHardMachine Learning EngineerAnalyze matrix multiplication complexity
This question evaluates understanding of time and space complexity for matrix multiplication and memory accounting in ML workloads, testing competency in asymptotic analysis, linear algebra operations, and resource estimation.
Software Engineering FundamentalsTechnical Screen - OpenAIHardData ScientistAssess free-month promotion impact
This question tests your ability to design a rigorous experimentation framework to measure the causal impact of a promotional offer while accounting for self-selection bias and long-term churn behavior.
Analytics & ExperimentationTechnical ScreenA/B TestingCausal InferenceProduct Analytics - OpenAIMediumSoftware EngineerAt-Least-Once Job Queue with Timeouts, Retries, a Dead-Letter Queue and Dependencies
Implement an in-memory job queue with enqueue, reserve, pass, fail and a dead-letter queue, where reservations that outlive their timeout are redelivered for at-least-once delivery, then extend enqueue with job dependencies. It tests state-machine design, timeout tracking, retry accounting and dependency handling.
Software Engineering FundamentalsTechnical Screen - OpenAIMediumAndroid EngineerBuild a Compose Rating Card
Tests proficiency in modern Android development with Kotlin and Jetpack Compose, focusing on unidirectional data flow (UDF), custom stateless composables, ViewModel state management, and asynchronous submission handling.
Coding & AlgorithmsTechnical Screenandroidjetpack-composekotlin - OpenAIMediumFrontend EngineerBuild a Streaming Chat Input
This question evaluates frontend engineering skills including streaming data handling, incremental token-by-token rendering, asynchronous state management, and preserving UI responsiveness during long-running API streams.
Coding & AlgorithmsTechnical Screen - OpenAIHardMachine Learning EngineerBuild and troubleshoot image classification and backprop
Build and troubleshoot image classification and backprop 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 - OpenAIHardSoftware EngineerCheapest Daily Shirt Colors With No Color Repeated on Consecutive Days
Pick a blue, green or red shirt for each of n days so that the total price is as low as possible while never buying the same color on two consecutive days, and return the chosen color sequence. It tests dynamic programming over a small state space and reconstruction of the optimal choices.
Coding & AlgorithmsTechnical Screen - OpenAIHardSoftware EngineerCheapest Paint-Booth Color Schedule With No Repeat on Consecutive Cars
Choose one of three paint colors for each car on an assembly line to minimize total cost, never repeating a color on consecutive cars, and return the exact color chosen for each car.
Coding & AlgorithmsTechnical Screendynamic-programmingreconstructiongreedy-vs-dp - OpenAIMediumSoftware EngineerClarify a Machine Topology Coding Task
Implement and formalize an API to analyze distributed accelerator interconnect topologies, separating summary metrics from component-level graph structures and validating topologies like rings, cliques, and lines.
Software Engineering FundamentalsTechnical Screengraph-algorithmsbfsapi-design - OpenAIHardSoftware EngineerCompute Infection Time in a Grid
This problem tests your ability to model state propagation in a 2D grid using Breadth-First Search (BFS) to determine the time required for a dynamic condition to cover all reachable nodes.
Coding & AlgorithmsTechnical ScreenBFSGraphSimulation - OpenAIMediumSoftware EngineerCompute Plant Infection Stabilization
This problem tests your ability to simulate state propagation across a grid over discrete time steps and determine the minimum number of interventions required to contain an infection.
Coding & AlgorithmsTechnical ScreenSimulationGridBacktracking - OpenAIMediumSoftware EngineerCompute Plant Infection Time
This problem tests your ability to model a grid-based propagation process using Breadth-First Search (BFS), specifically handling multi-source initialization and obstacle avoidance.
Coding & AlgorithmsTechnical ScreenBFSGraphGrid - OpenAIHardData ScientistCompute churn with re-subscriptions
This problem tests your ability to reconstruct user subscription states over time from raw event logs, specifically identifying active status to calculate monthly churn rates.
Data Manipulation (SQL/Python)Technical ScreenSQLData ManipulationTime-series - OpenAIEasyData ScientistCompute signup rate and retention from raw logs
This question evaluates the ability to manipulate event-level logs to compute cohort metrics—specifically signup rate and D30 retention—testing proficiency in SQL/Python, event-time windowing, joins, aggregation, and understanding of upstream logging semantics.
Data Manipulation (SQL/Python)Technical Screen - OpenAIHardSoftware EngineerContiguous Extent Allocator for a Storage Volume, Then O(log n)
Build an allocator that hands out contiguous runs of blocks on a fixed-size volume and takes them back with coalescing, first with a simple sorted list and then with logarithmic reserve and release. Tests splitting/merging logic, choosing indexes for size and neighbor lookups, and keeping them consistent.
Software Engineering FundamentalsTechnical Screenallocatordata-structuresbalanced-bst - OpenAIMediumBackend EngineerConvert IPv4 Ranges to CIDR Blocks
This question evaluates understanding of IPv4 addressing, binary/bitwise manipulation, and the ability to summarize consecutive address ranges into minimal CIDR blocks, and belongs to the Coding & Algorithms domain for backend engineering roles.
Coding & AlgorithmsTechnical Screen - OpenAIMediumSoftware EngineerCount Connected Friend Groups
Determine the number of connected components in an undirected graph represented as an adjacency matrix using graph traversal or disjoint set union.
Coding & AlgorithmsTechnical Screengraphdfsbfs - OpenAIMediumSoftware EngineerCount Machines and Recover a Distributed Tree Topology
Implement a distributed tree discovery and aggregation protocol that counts reachable machines, recovers edges, handles network retries, detects unreachable subtrees, and validates tree invariants.
Software Engineering FundamentalsTechnical ScreenDistributed SystemsTreeGraph Traversal - OpenAIMediumMachine Learning EngineerDebug MiniGPT and Backpropagate Matmul
This question evaluates proficiency in debugging PyTorch transformer implementations, reasoning about attention mechanics and causal masking, implementing key–value caching for autoregressive decoding, and constructing provably correct backward passes for matrix multiplications and custom autograd.Functions.
Machine LearningTechnical Screen - OpenAIMediumMachine Learning EngineerDebug a Concurrent Job Scheduler
This question evaluates debugging and hardening skills for concurrent systems, focusing on detection and reasoning about race conditions, deadlocks, lock-contention hotspots, state-machine invariant enforcement, rate-limiting, retry semantics, and instrumentation for per-job metrics in a job scheduler.
Software Engineering FundamentalsTechnical Screen
OneMain Financial
OpenAI