Skip to contentSkip to footer

Scientist Interviews

Salaries
Interviews

Scientist Interview Questions

"The questions you are asked in an interview for a position as a scientist will depend greatly on field of science you intend to work in. Generally, interviewers will be interested in your formal education, field of study and specialization, work, internship, and research experience, scientific writing skills, and interest in the subject matter. Expect to be asked technical questions that pertain to the knowledge needed to perform the duties of the job. While there are some positions open to scientists who possess associates' or bachelors' degrees, most jobs will require you to have at least a masters' degree with the majority requiring you to have a doctorate."

54,359 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 54,359 interview questions and reports from Scientist interviews. Prepare for your interview. Get hired. Love your job.