Google Data Scientist Interview Questions
& Process
Real candidates share what happened, how many rounds they had,
and how the experience turned out.
Based on 133 interview experiences · FREE TO READ
Candidate interview experiences
First-hand accounts from people who interviewed at Google.
Google
Data Scientist
Senior Easy Neutral experience Decline offer 2 rounds Same day
Interview process
Technical screen
Interview formats
Behavioral Technical Coding System Design
I successfully cleared the first round of the interview, where I talked about my experience and technical background in detail. During this round, I described how I built a custom TensorFlow-based library to deploy and use machine learning models in real-time environments, with a focus on performance and scalability. On the same day, I also attended the DSA round and solved the given problems efficiently, showing strong problem-solving skills and a solid understanding of data structures and algorithms. Overall, the interview went well and showcased my hands-on experience with both system design and algorithmic thinking.
Confirmed questions1 question
- how would you create a custom tensor flow library?
Google
Data Scientist
Lead Very difficult Neutral experience No offer 2 rounds About 7 days
Interview process
Recruiter call Technical screen
Interview formats
Behavioral Technical
Recruiter call about 35m - friendly and straightforward, more like fact-finding. Technical call about 60m - pretty difficult and hard to explain things properly over the phone. I wasn’t moved forward to on-sites - the recruiter mentioned on the call that the next steps would have been multiple on-site interviews. Mostly questions were about my experience and “tell me about when you have worked on ...” type things.
Confirmed questions1 question
- Tell me about a time you worked on an end-to-end ML model and getting it deployed.
Google
Data Scientist
Mid Average Positive experience No offer 6 rounds
Interview process
Recruiter call Technical screen Onsite
Interview formats
Behavioral Technical Coding Case Other
1 round of recruiter phone screen, 4 rounds of technical interviews covering the following areas: SQL: Focused on data querying, data manipulation, and optimization skills. A/B Testing: Assessing your ability to design experiments, perform statistical analysis, and interpret results. Modeling: Evaluating your knowledge and practical skills in machine learning or statistical modeling (e.g., feature engineering, model selection, performance evaluation). 1 round of Googliness & Leadership interview.
Confirmed questions1 question
- SQL A/B testing case study.
Google Data Scientist Interview Questions
Quoted word for word from Google interview reports.
“Given 3 tosses of a fair coin are Heads what is the probability that the 4th coin toss is a Head”
Read reports →“Q6. In what way does multicollinearity impact prediction accuracy versus how it impacts the trustworthiness of model weights?”
Read reports →“If the individual mean of two groups is going up, can the combined pooled mean still go down? If yes, then explain how?”
Read reports →“how to sample 100 values between 2 and 3 from standard normal distribution, and how would you optimise it so you get exactly 100 samples instead of relying on a long accept/reject loop”
Read reports →“Given a sample of just 1 point, design a test to check if the mean is 1 or not.”
Read report →“Is it good or bad if you use bootstrapping on samples to make your sample size bigger?”
Read report →“How are 2 time series different from each other, write a code for it.”
Read report →“📈 4. Regression & Multicollinearity Q4. How would you define multicollinearity in linear regression?”
Read report →“Explain how a neural network works. Why do you think a sigmoid function is more useful as an initialised function or otherwise?”
Read report →Formats, difficulty and experience
Across all 133 Google interview reports.