Skip to contentSkip to footer

Data Scientist Interviews

Salaries
Interviews

Data Scientist Interview Questions

In a data scientist interview, expect employers to ask questions that assess your data modeling, problem-solving, and programming skills. Be prepared to answer general questions that test your knowledge of statistics and data science. You should also be ready to answer open-ended questions that test your creativity, communication skills, and formal education in data modeling and programming.

Top Data Scientist Interview Questions & How to Answer

Question 1

Question #1: Which data modeling techniques do you prefer and why?

How to answer
How to answer: Turning data into understandable and actionable information is a critical part of the data scientist's job. This question allows employers to understand your data modeling skills and background. List and discuss your preferred data modeling techniques, including benefits such as ease of use, flexibility, etc.
Question 2

Question #2: How would you detect bogus Instagram accounts used for scamming consumers?

How to answer
How to answer: Questions like this one allow an employer to test your problem-solving skills. When answering open-ended questions such as these, feel free to ask clarifying questions and use whiteboards to demonstrate your coding and diagramming skills. Share your thought process as you work through the problem.
Question 3

Question #3: Describe circumstances that require a list, tuple, or set in Python.

How to answer
How to answer: Interviewers will use questions such as this one to test your Python programming skills. Review Python basics such as lists, tuples, and sets before your interview. You should be able to explain when and how each tool is used by data scientists.

54,338 data scientist interview questions shared by candidates

Where does Deep Learning offer advantage compared to SVMs? Is the cost function of a DNN model convex? What about for SVM? Tell me about how you have implemented a research paper (mentioned in my resume) Basic questions about linear and logistic regressions - about their assumptions, advantages etc Overall, the questions weren't too deep.
avatar

Senior Data Scientist

Interviewed at Ericsson

4★
Oct 14, 2018

Where does Deep Learning offer advantage compared to SVMs? Is the cost function of a DNN model convex? What about for SVM? Tell me about how you have implemented a research paper (mentioned in my resume) Basic questions about linear and logistic regressions - about their assumptions, advantages etc Overall, the questions weren't too deep.

1. What's the relationship between PCA and k-means clustering? 2. What are the requirements for a matrix to represent a kernel? What happens if we run SVM using a 'kernel' that does not satisfy these requirements? 3. Problems using Python lists and dictionaries 4. SQL joins, aggregates (count, sum, avg), and cases 5. If you were given a dataset with [X] features (may be numerical, categorial, etc.) and you want to build a model (to determine fraudulent transactions, say), how would you determine which features are best to use in the model?
avatar

Data Scientist

Interviewed at Palo Alto Networks

3.8★
Apr 27, 2019

1. What's the relationship between PCA and k-means clustering? 2. What are the requirements for a matrix to represent a kernel? What happens if we run SVM using a 'kernel' that does not satisfy these requirements? 3. Problems using Python lists and dictionaries 4. SQL joins, aggregates (count, sum, avg), and cases 5. If you were given a dataset with [X] features (may be numerical, categorial, etc.) and you want to build a model (to determine fraudulent transactions, say), how would you determine which features are best to use in the model?

Viewing 1461 - 1470 interview questions

Glassdoor has 54,338 interview questions and reports from Data scientist interviews. Prepare for your interview. Get hired. Love your job.