Amazon Data Scientist | June 2020

?
AnonymousApril 5, 2021, 01:12 AM
Upvote
1
Downvote

Company: Amazon

Position: Data Scientist

Location: nan

Level: Mid-level

Outcome: NA

How was the interview process? What was it like?

I had three rounds of technical interviews. First, two interviewers were pretty chill in their approach.  Most of the questions were around data science modeling and scenarios. Few of the questions that I can recall are: 
1. Metrics for forecasting? RMSE, MAPE, SMAPE which one to use when 
2. How to calculate the confidence interval of random forest error? 
3. You put a model into production, how will you know if it needs to be remodeled? this was at the end of the second interview, and the interviewer wasn't happy with my response that we need to measure data drift. His answer was A/B testing. The 3rd interviewer asked the same question again and I responded to him that we can do A/B testing. He wasn't convinced lol. His answer was that we can build ensemble of models and test if the ensemble is better than the one in production and that's how they do it on Kaggle :D
Overall it was pretty fun to interview especially the last one :P The first two interviews were taken by L5, L6 level guys and the last one was L7

What technical questions were asked?

Modeling and ML Knowledge, Python / Pandas, Statistics

What was one of your solutions?

Thought about KL divergence or using the distance between the theoretical quantiles and empirical quantiles. My interviewer was funny in the sense that he didn't accept q/q plot or statistical tests for normality as robust methods to check the normal distribution of data. Finally, I coded the distance between empirical and theoretical quantiles ( which is in a way what q/q plot will show) but his responses was "meh".  It will be interesting to see what you guys think about this problem. Also, at the end of the interview(like 5 minutes remaining) I was asked to write a solver for a linear regression problem with 10Million rows and 100 columns.  I started with a gradient descent code, the interviewer asked if there are any other solvers. Normal equations but it won't scale, to which he seemed satisfied.

Amazon
Data Scientist
I
Ichimoku
Top comment
Lorem ipsum dolor sit amet, consectetur adipiscing elit
sed do eiusmod tempor incididunt ut labore et dolore magna aliqua.

Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur.

Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
F
Fibonacci
Sit voluptatem accusantium doloremque laudantium, totam rem aperiam, eaque ipsa quae ab illo inventore veritatis et

quasi architecto beatae vitae dicta sunt explicabo. Nemo enim ipsam voluptatem quia voluptas sit aspernatur aut odit aut fugit, sed quia consequuntur magni dolores eos qui ratione voluptatem sequi nesciunt. Neque porro quisquam est
I
Ichimoku
Sed ut perspiciatis unde omnis iste natus error sit voluptatem accusantium doloremque laudantium 😀
F
Fibonacci
Sit voluptatem accusantium doloremque laudantium, totam rem aperiam, eaque ipsa quae ab illo inventore veritatis et

quasi architecto beatae vitae dicta sunt explicabo. Nemo enim ipsam voluptatem quia voluptas sit aspernatur aut odit aut fugit, sed quia consequuntur magni dolores eos qui ratione voluptatem sequi nesciunt. Neque porro quisquam est

Interview Guide

View the full Amazon Data Scientist interview guide

Related posts

Amazon Manager Business Intelligence Manager | June 2021
UBS Senior Quantitative Analyst | April 2021
Bristol Myers Squibb Senior Research Scientist | March 2021
Amazon Entry Level Cloud Support Engineer Analytics | August 2022
There's so much more to Interview Query! Sign up to access hundreds of interview questions, expert coaching and a flourishing data science community.