Top 50
Prep plan
Updated weekly · Last refresh Oct 17

Google Research Scientist Interview Questions

The questions to prepare for a Google Research Scientist interview. Questions from real interview reports rank first. Updated daily.

50questions
~8htotal time
Track your progressSign up free to work through all 50 questions and resume where you left off.
Start practicing free →
1
CodingStart here. 14 questions · ~141 min
Trees and Dynamic ProgrammingHard
Practice

Use tree dynamic programming to find the maximum sum of non-adjacent nodes in a binary tree.

Dynamic ProgrammingdfsAlgorithmsGoogle
Streaming Median with Low MemoryHard
Practice

Calculate streaming bid medians with two heaps while keeping memory proportional to the observed bids.

aggregationArraysSortingGoogle
More Coding questions with a free account
2
Statistics & Probability16 questions · ~161 min
Robust A/B Testing for ScaleHard

Design a robust Google A/B test covering randomization, power, inference, monitoring, and rollout risks.

ExperimentationPower AnalysisSample SizeGoogle
Separate Signal From Noise and BiasHard

Design a rigorous statistical workflow to distinguish real effects from noise and bias in large-scale Google datasets.

ExperimentationBiasCorrelationGoogle
More Statistics & Probability questions with a free account

Sign up to see every question

Create a free account to unlock this list and practice real interview questions.

Get my prep plan
3
Machine Learning15 questions · ~151 min
Transformer vs CNN vs RNN ChoiceHard

Select and validate a Transformer, CNN, or RNN by matching architecture to sequence structure, data scale, latency, and evaluation needs.

Neural NetworksBias-Variance TradeoffDeep LearningGoogle
Trade-offs Between Transformers, CNNs, and RNNsHard

Compare Transformers, CNNs, and RNNs across sequence modeling contexts, considering accuracy, compute, latency, data, and deployment trade-offs.

Neural Networksmodel architectureBias-Variance TradeoffGoogle
More Machine Learning questions with a free account
4
Model Evaluation4 questions · ~40 min
Right Metrics and EvaluationHard

Explain how to validate data quality and select metrics that reflect a model's real-world objectives and risks.

evaluation metricsPrecisionAccuracyGoogle
Scientific Integrity in EvaluationHard

Explain how to design rigorous, reproducible, unbiased model evaluations and handle data responsibly at Google.

Evaluation TechniquesConfusion MatrixCalibrationGoogle
More Model Evaluation questions with a free account
5
More topics1 question · ~10 min
More questions with a free account
The finish line: interview-readyComplete all 50 questions to finish this plan.