Top 50
Prep plan
Updated weekly · Last refresh Oct 17

Google DeepMind Research Engineer Interview Questions

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

50questions
~7htotal time
Track your progressSign up free to work through all 50 questions and resume where you left off.
Start practicing free →
1
CodingStart here. 8 questions · ~69 min
2
Model Evaluation7 questions · ~60 min
Evaluate Model CalibrationHard

How to tell whether a model's predicted probabilities are well calibrated, and what the business impact is.

Log LossCalibrationAUC-ROCGoogle DeepMind
More Model Evaluation 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
System Design8 questions · ~69 min
Online vs Batch Model ServingMedium

Compare batch and online serving for an ML ranking system, including freshness, latency, cost, and operational complexity.

Feature StoreRetrievalModel ServingGoogle DeepMind
More System Design questions with a free account
4
NLP4 questions · ~34 min
Explain TF-IDF for Text FeaturesEasy

Explain TF-IDF and where it helps in text classification and search.

Text ClassificationTF-IDFTokenizationGoogle DeepMind
Explain Word EmbeddingsEasy

Explain how word embeddings represent words as dense vectors and why they help NLP models capture meaning.

Language ModelsText ClassificationFeature EngineeringGoogle DeepMind
More NLP questions with a free account
5
Machine Learning8 questions · ~69 min
Feature Engineering for Supervised ModelsEasy

Explain feature engineering and why transforming raw inputs can materially improve supervised model performance.

Cross-ValidationFeature EngineeringSupervised LearningGoogle DeepMind
More Machine Learning questions with a free account
6
Statistics & Probability8 questions · ~69 min
Interpreting P Values in TestingEasy

Explain what a p-value means in hypothesis testing and how it relates to statistical significance.

Hypothesis TestingStatistical SignificanceP-ValuesGoogle DeepMind
More Statistics & Probability questions with a free account
7
More topics7 questions · ~60 min
Batch vs Streaming Data ProcessingEasy

Compare batch and streaming data processing, including when each fits best in a pipeline.

Stream ProcessingETLBatch ProcessingGoogle DeepMind
More questions with a free account
The finish line: interview-readyComplete all 50 questions to finish this plan.