KL divergence is a symmetric dissimilarity function between distributions?
Data Science Manager Interview Questions
331 data science manager interview questions shared by candidates
Why this company? Adyens Formula - memorize it Previous technical cases in similar contexts.
Linear regression, problem solving analytics
Zero technical question. Current company's insider information. Salary negotiation.
How would you pick the "winner" of the A/B test?
How would you go about analyzing user engagement with a new feature and identifying areas for improvement?
Tell me something about yourself - describe yourself.
How do you tell clients that you need more data?
Tech screen / SQL - Find the top 10 users with most friends - Count the content for ‘shares’ and ‘comments’ amongst 2 users over the course of a year Onsite Case Analysis 1 - How would you assess the success of News Feed if you were to show content from family and friends primarily? - How would you define people you know in person and show content from those? - How would you measure the downside of showing primarily content from family and friends? - How about notifications sent to users – how would you define which notifications to send to users based on their close family etc? - How would you measure if the notifications you are sending are working? - How would you go about testing it? - How would you disambiguate results from an A/B test? - What kind of metrics would you track? Product Leadership – Cross-functional Partnership - Did you even have to push back product partners to collect more data to define/prioritize work? Case Analysis 2 - How would you measure the success of using smaller font in News Feed for text-based, video, photo content? - Say that the number of comments is decreasing but the session time is increasing? How would you assess whether that is good/bad - Say that decreasing the font size worked better for users in the US but not for folks in Thailand People Leadership - Questions about growing a team, etc.
SQL: count, joins, avg on FB data Case: If FB is going to launch a new feature, what should you do? How do you apply A/B testing? What would be the sample size? What metrics should you check?
Viewing 121 - 130 interview questions