FORESIGHT(2023): Summer Internship at Amex| Sagar Kumar Karn|

1) Brief Introduction

I’m Sagar Kumar Karn, a final-year undergraduate student of Economics at IIT Kharagpur. I am currently working at American Express as an Analyst Intern. I am a boarder of Patel Hall of Residence.

2) How did you get into Amex? What was the selection procedure?

I started exploring Data Science in my second year and became quite interested in this field. I pursued two research internships in this domain in my third year. One of them was at the Insitute of Technology and Business, Czech Republic, and another at the Centre for Analytical Finance. These internships helped me to gain a further understanding of this field.

Now coming to the selection procedure, the test comprised of three sections namely Aptitude, Machine Learning, and Case Study. The first section had 20 MCQ questions on aptitude and basic Prob stats and had to be solved in 40 minutes. The questions were of CAT level and included puzzles, probability and statistics, etc.

The second section had 10 questions on ML and had to be done in 20 minutes. To get through this section you need to have a clear understanding of various ML algorithms and models. The questions were centered around Decision Trees, XGBoost, black box models, k nearest neighbors, etc.

The third section is a case study section where you are given a case and you are required to give an answer to a question. To arrive at the answer you have 50 questions at your disposal and you can unlock at max 25 of them. The fewer questions you unlock to get to the correct answer, the better your score is. You can attempt either or both of the second and third sections. The best of the two is taken into account. The first section is compulsory.

Now coming to the interview, there were two interview rounds. The first round was mostly based on my knowledge of machine learning and deep learning. One of the questions in the first round was on how I was asked about how XGBoost works and what is boosting. The second round basically revolved around my resume and my previous internships and projects. The interviewer asked me to explain one of my projects and then after I was done, asked various cross questions on the same. It is therefore important to be thorough with the internships and projects that you mention in your resume because they can easily see through if have just written it but you don’t have a basic understanding of the project.

3) How to prepare for them?

To prepare for Probability and Statistics, “50 Challenging Problems in Probability” and “Heard on the Street” are great books. You can practice for the aptitude section from Brainstellar.

For machine learning and deep learning, it is important to have a basic understanding of different algorithms and models. One of the best resources to start off with Machine Learning is the course by Andrew NG for machine learning. For deep learning, Deep Learning Specialization on Coursera is a good resource. Other than that Stanford University courses such as CS230(Deep Learning), CS231n(Computer Vision), CS224n(Natural Language Processing), etc are also great courses to follow.

Once you are done with the theory do some self-projects to test your understanding and practical implementation of what you have learned.

4) What difficulties did you face while preparing for this Company/Profile? How did you overcome this problem?

One of the basic difficulties I faced was that I did not take enough time to revise all the stuff before the interviews. I was preparing for both SDE roles and Analytics roles and had devoted more time to DSA, so had quite less time to revise Data Science topics. My advice to people prepping for both profiles is to keep enough time for the revision of these topics.

5) According to you, who should ideally apply for this job?

If you have an interest in Data Science and want to work on real-life problems involving data science in the credit card industry then this is one of the best places to work. Work that you do here would actually add value to the company. The work here is more on the implementation and analytical side so folks who are really interested in the implementation of ML algorithms should definitely go for it.

6) Any specific advice you want to give to the junta sitting for internships this year?

My advice to my juniors would be to decide the roles that they prepare for according to their interests. Instead of following the rat race develop your skill set in the particular field you actually like and want to work on. Don’t just get into coding if you don’t actually like coding. Explore various fields and decide according.

Another piece of advice I want to give is to remain positive and confident throughout the process. It is a cruel process, a bit dependent of luck and a lot of other factors that you cannot control. But always remember CDC internship is just a way for you to get an internship. It is certainly not the end of the world. There are ample opportunities out there. If you are not getting an internship through CDC then you shouldn’t feel disheartened. Keep working hard and expanding your skill set. Keep looking out for off-campus internships and competitions.

Lastly, I would reiterate that you should decide whether you “want to get an intern” or you “want to get an intern in a field that you are interested in”. Depending on what is your answer prepare accordingly. If you want to prepare for more than one profile then prioritize accordingly and manage time effectively. Also, be confident during the interviews, and if you don’t know the answer to any question just politely say that you don’t know instead of meandering here and there and wasting time. Be honest in the interview and don’t mention something in your CV that you haven’t actually done.

7) What are some of the major points you think would be valid to mention in your CV while targeting this profile? (any specific suggestions you would like to make?)

When preparing for interviews in Data Analytics, do include one or two projects in this domain. In interviews for Data profiles, your CV plays quite an important role. Having projects in ML or DL would surely be a plus point on your CV and you would be able to drive your interview around the projects you have done. Also when mentioning an intern or a self project in your CV, the impact created (if any) and the improvement from a previous approach(if any) should be written.