Data Science interview Experience with Amex

Hi Everyone, It’s good to share the interview experience no matter whether it’s a success or a failure

Intro (5min)

It started with my CV walk-through to thoroughly understand the projects that I worked on and their impact on the business further.

1st: Project discussion (30 min)

1. Start discussing one good project that you end-to-end?
2. how did you do the data preparation and data cleaning? what are all steps are involved?
3. How did you perform the feature engineering?
4. what kind of feature selection technique is used to reduce the variables?
5. How are you calculating the information value?
6. What kind of sampling technique has been used?
7. what is the difference between Xgboost and random forest?
8. what are the metrics that are used to measure the performance?
9. How did you perform the validation?
10. how did you handle the over-fitting of the problem?

2nd: Case study discussion (3–5 min for thinking, rest 10 min)

Get sasi’s stories in your inbox

Join Medium for free to get updates from this writer.

Remember me for faster sign in

There we need to prepare an approach for an ML problem which can be an ongoing one or a new one entirely.

3rd: Questions on Python and SQL (10 min)

1. what are the different packages that you have used?
2. Which package is to be used for a particular situation?
3. how to create new columns based on the existing columns with conditions provided?
4. Can you repeat the same in SQL and what is the query?

Questions to the interviewer and about the company from my end (5min)