Waymo interview questions

Showing 61–90 of 90 results

• How predict vehicles’ turn direction at intersection?Easy

  This problem tests the ability to design an end-to-end machine learning pipeline for vehicle intent prediction, focusing on feature engineering, sequence modeling, and temporal validation.

  WaymoData ScientistMachine LearningOnsite
• How validate a driving simulation is realistic?Easy

  This question tests the ability to design an evaluation framework to compare simulated vehicle behavior against real-world sensor and trajectory logs to ensure high-fidelity simulation.

  WaymoData ScientistAnalytics & ExperimentationOnsite
• Implement Fibonacci generator, balanced BST, frozensetMedium

  This problem tests your understanding of Python generators for sequence generation, the mechanics of tree rotations in self-balancing data structures, and the requirements for implementing immutable, hashable collections.

  WaymoSoftware EngineerCoding & AlgorithmsTechnical Screen
• Implement K-means and handle train-inference mismatchEasy

  This question evaluates understanding and hands-on competency in unsupervised clustering (K-means objective, alternating optimization, initialization strategies, empty-cluster handling, stopping criteria, and computational complexity) and sequence modeling for multi-agent trajectory prediction (input representation, attention/GNN architectures, deterministic vs probabilistic outputs, evaluation metrics, and exposure bias from autoregressive training). It is commonly asked to probe both conceptual understanding and practical implementation skills in the Machine Learning domain, testing optimization and scalability trade-offs as well as model-design and training/inference mismatch considerations at a mix of conceptual and practical application abstraction levels.

  WaymoData ScientistMachine LearningTechnical Screen
• Implement Safe Average FunctionMedium

  This question evaluates proficiency with basic programming fundamentals including numeric computation, handling edge cases such as empty inputs, input immutability, and data type consistency.

  WaymoData ScientistCoding & AlgorithmsOnline Assessment
• Implement a Safe Average Function in PythonMedium

  Implement a Python average function that returns zero for an empty list and explain its time and space complexity. Clarify numeric types, missing values, NaN behavior, streaming alternatives, precision concerns, and the tests needed for a reliable data utility.

  WaymoData ScientistData Manipulation (SQL/Python)Online Assessment
• Implement an ObjectTracker That Merges Observations From Two Perception SystemsMedium

  Two perception systems track the same objects under different IDs. Implement an ObjectTracker with addLink, addObservation and getHistory that returns an object's observations from both systems in timestamp order, testing identity grouping and history merging.

  WaymoSoftware EngineerSoftware Engineering FundamentalsOnsite
• Implement nested template string substitutionEasy

  This problem tests your ability to handle recursive data structures and string parsing by expanding nested placeholders within a template string while identifying potential circular dependencies.

  WaymoData ScientistCoding & AlgorithmsOnsite
• In-Place Rectangle Copy Within a 2D Array Using O(1) Extra MemoryMedium

  Copy a rectangular block of a 2D array to another position in the same array, in place with O(1) extra memory, where the source and destination blocks may overlap. Tests choosing a safe copy direction so that no value is overwritten before it is read.

  WaymoSoftware EngineerCoding & AlgorithmsOnsite
• Insert +, -, * and Parentheses Between Ordered Numbers to Reach a TargetMedium

  Given numbers in a fixed order and a target, insert +, - or * between them and add parentheses to reach the target, returning the lexicographically smallest fully parenthesized expression or an empty string. Starts with three numbers and generalizes to n, testing exhaustive search over expression trees.

  WaymoSoftware EngineerCoding & AlgorithmsOnsite
• Interpret a Smart Wait Launch with Conflicting MetricsMedium

  Interpret a product launch where conservative wait estimates reduce conversion while observed pickup time also falls. Diagnose selection effects and alternative causes, define customer and fleet metrics with consistent denominators, and propose an experiment that can separate display accuracy from operational improvement.

  WaymoData ScientistAnalytics & ExperimentationOnline Assessment
• Is the Segment From 0 to 50 Fully Contaminated After Each Landing Point?Medium

  Points land one at a time on the segment from 0 to 50, each contaminating an interval of length 1 around it; after each landing, report whether the whole segment is contaminated. Tests interval union maintenance, exact boundary handling and efficient updates.

  WaymoSite Reliability EngineerCoding & AlgorithmsOnsite
• Lowest Horizontal Cut That Splits Square Cakes Into Equal Total AreasMedium

  Square cakes sit on a table with sides parallel to its edges; find the lowest horizontal cut line that splits the total cake area equally, returned as an exact fraction. Tests sweeping over heights, piecewise-linear area accumulation and exact arithmetic.

  WaymoSoftware EngineerCoding & AlgorithmsOnsite
• Merge overlapping intervals per group in pandasEasy

  This problem tests your ability to handle temporal data by merging overlapping time intervals within distinct groups using pandas and custom logic.

  WaymoData ScientistData Manipulation (SQL/Python)Onsite
• Minimize Total Grid Distance to Every BuildingMedium

  Find an empty land cell in a 2D grid that minimizes the sum of shortest travel distances to all buildings without traversing through obstacles or other buildings.

  WaymoSoftware EngineerCoding & AlgorithmsOnsite
• Model Rare Events with a Log Link and Handle Zero CountsMedium

  Explain rare-event log-link models, exposure offsets, zero counts, overdispersion, and the difference between counts and binary outcomes.

  WaymoData ScientistStatistics & MathTechnical Screen
• Model wins-until-failure and expected future winsEasy

  This question tests your ability to model geometric distributions and apply maximum likelihood estimation (MLE) to censored data.

  WaymoData ScientistStatistics & MathOnsite
• Optimize Tensor Runtime KernelsMedium

  This question evaluates competency in optimizing machine learning framework runtimes for accelerators, covering profiling and performance analysis of tensor kernels, memory layout design, operator scheduling and kernel fusion while weighing trade-offs among throughput, latency, memory usage, numerical correctness, and maintainability.

  WaymoMachine Learning EngineerSoftware Engineering FundamentalsTechnical Screen
• Parse a Raw CSV String Into a Structure for Downstream Teams, Handling Corrupted RowsMedium

  Parse a raw CSV string from scratch into a data structure other engineering teams can consume, with the prompt given verbally and only one sample record. The follow-up handles corrupted rows where the header is trusted but rows may be missing fields, testing tokenizing, API design and error reporting.

  WaymoSoftware EngineerSoftware Engineering FundamentalsOnsite
• Respond to On-Call Pressure, Feedback, and Team ConflictMedium

  This question assesses a candidate's ability to navigate high-pressure production incidents, manage interpersonal conflict under uncertainty, and drive long-term systemic improvements.

  WaymoSoftware EngineerBehavioral & LeadershipOnsite
• Return All Words Matching Each Prefix From a Word ListMedium

  Build a structure over a word list that answers prefix queries, returning every distinct word that starts with each prefix in lexicographic order. Tests prefix-tree design, deduplication, and answering many queries without rescanning the whole list.

  WaymoMachine Learning EngineerCoding & AlgorithmsOnsite
• Run Promise Tasks with Bounded ConcurrencyMedium

  Implement an asynchronous task runner in TypeScript that limits concurrent execution, handles synchronous throws and rejections gracefully, and returns settled results in the original task order.

  WaymoFrontend EngineerSoftware Engineering FundamentalsOnsite
• SQL: Average Ride Rating in Users' First and Third Active Months by CityMedium

  Write a PostgreSQL query that compares average ride ratings in each user's first and third active months, grouped by city and averaged over individual rides. Tests ranking distinct active months per user, window functions, NULL ratings and weighting.

  WaymoData ScientistData Manipulation (SQL/Python)Onsite
• SQL: Users With No Rides in the Most Recent 7-Day Window, Including Non-RidersMedium

  Write a PostgreSQL query that lists users with no rides in the most recent 7-day window, including users who never rode. Tests anchoring a date window to the data, inclusive boundaries, and anti-join logic that keeps users without matching rides.

  WaymoData ScientistData Manipulation (SQL/Python)Onsite
• Sample Points Uniformly from a Unit DiskMedium

  Derive uniform random sampling inside a unit disk, explain the radial distribution, and compare a polar sampler with rejection sampling.

  WaymoData ScientistCoding & AlgorithmsTechnical Screen
• Sort the Results of Applying a Quadratic Function to a Sorted ArrayMedium

  Apply the quadratic function a*x*x + b*x + c to every element of a sorted integer array and return the results in sorted order, where a may be negative or zero. Tests reasoning about the parabola's shape and merging values from both ends in linear time.

  WaymoSoftware EngineerCoding & AlgorithmsOnsite
• Test Whether a Routing Experiment Reduced Pickup TimeMedium

  Analyze a routing A/B test in Python by filtering and joining user and ride data, then calculating a Welch t-test for pickup time. Interpret effect size and uncertainty, address repeated rides and post-treatment selection, and design stronger user-level or cluster-aware inference.

  WaymoData ScientistAnalytics & ExperimentationOnline Assessment
• Top 5 Most Efficient Vehicle ModelsMedium

  This problem tests your ability to aggregate data from multiple tables, calculate performance metrics, and handle common SQL pitfalls like division by zero and NULLs.

  WaymoData AnalystData Manipulation (SQL/Python)Technical Screen
• Validate a Parent-Child ForestMedium

  This problem tests graph theory fundamentals by requiring the validation of a forest structure, specifically checking for cycle-free properties and valid parent-child node connectivity constraints.

  WaymoSoftware EngineerCoding & AlgorithmsTechnical Screen
• Welch's t-Test p-Value for a Routing Experiment on Time to Pickup, and Its ValidityMedium

  Filter rides to one city and month, join users to their experiment variants, and compute a Welch's t-test p-value on time to pickup in Python. Then judge whether the routing test supports a conclusion and design a better experiment for a shared fleet.

  WaymoData ScientistAnalytics & ExperimentationOnsite