Are you a data scientist aspirant? Are you currently applying for data scientist positions? Do you have a data scientist interview coming up? Are you worried about the interview process?
If you have any of the above questions in mind, then you are in the right place. This article will help answer some of the questions you might have about the data scientist interview process and hopefully it will help calm your nerves and alleviate any anxieties before you begin your interview.
After going through a couple of data scientist interview processes, I would like to share my experiences with aspiring data scientists. Hopefully, they’ll learn something from my experiences that could help them to be better prepared for the interview process.
Step 1: Initial Contact via LinkedIn from a Recruiter
All the 2 data scientists interviews that I’ve participated in started with a Recruiter contacting me via LinkedIn. The initial message looks like this:
“Hey, Benjamin! I came across your profile on LinkedIn and wanted to reach out to see if you would be open to new opportunities? One of our clients in the Tulsa, OK area that is currently a leader in the web-based monitoring and field automation services for oil and gas is looking for a Data Scientist to join their team! I thought it lined up well with your experience and wanted to make sure I ran it by you! Would there be a good time I could give you a call today to talk more about this role? Thanks and hope to speak with you soon!”
After this initial contact, I exchanged cell phone number and email address with the Recruiter and a Skype interview was scheduled with the Recruiter.
Lessons Learned: You can actually get a job via LinkedIn. So if you are an aspiring data scientist, make sure your LinkedIn profile is up to date, and you also want to let recruiters know that you are actively searching for a data scientist role. You may also join different data science groups on LinkedIn or follow top data science online publications such as Towards AI.
Step 2: Skype Interview with Recruiter
At the Skype interview, the Recruiter asked about my background and what kind of data science projects I’ve worked on in the past. Basically simply trying to figure out if my background and interests line up with the data scientist role being advertised. We also discussed at length about the following:
(i) The data scientist position
(ii) The job location
(iii) Required qualifications
(iv) Job expectations
(v) Work environment
(vi) Approximate pay and benefits
(vii) If I’ll be needing sponsorship for employment
Lessons Learned: Always be professional during each phase of the interview. Before the actual Skype interview, it’s important to check all your devices and internet connection to make sure all systems are functioning correctly. Be sure to show up on time, and always dress properly. During the interview, it’s always good to ask as many questions about the position as possible.
Step 3: Automated Video Interview
It’s now 2 weeks into the interview process. This time around I was contacted by the director of data science at the company. He asked me to complete an automated video interview. His message read like this:
“Dear Benjamin,Thank you for your interest in a career at eLynx Technologies LLC. We have reviewed your application for Data Scientist — Analytics and we would like to invite you to complete a video interview. You can take the interview at any time with the use of a computer with a webcam. The entire interview will take approximately 10:00 minutes and must be submitted by April 02, 2019.”
This step went successfully.
Lessons Learned: Here are some tips for video automated interviews:
(i) It’s important you have a broad knowledge about the company and the role you are being considered for. So do your research about the company as you may have to tell them what you know about the company in a nutshell. A good source of information would be the company’s website.
(ii) Be mindful of your background/surroundings (what is in view of the camera) during the interview. Make sure you are in a quiet place that is free from distractions.
(iii) Dress to impress: Always recommended to dress professionally all throughout.
(iv) Keep eye contact with the camera like it’s the eyes of the person you’re interviewing with.
Step 4: Take Home Data Science Challenge Problem
It’s now the 3rd week into the interview process. At this step, I was assigned a take-home data science challenge problem. This is generally a data science problem e.g machine learning model, linear regression, classification problem, time series analysis, etc. Generally, they provide you with project directions and the data set. If you are fortunate, they may provide a small data set that is clean and stored in a comma-separated value (CSV) file format. That way you don’t have to worry about mining the data and transforming it into a form suitable for analysis. For the couple of interviews I’ve had, I worked with 2 types of data sets, one had 160 observations (rows) while the other had 50,000 observations.
A sample take-home coding exercise could look like this:
Coding Exercise for the Data Scientist Position (Take Home)
This coding exercise should be performed in python (which is the programming language used by the team). You are free to use the internet and any other libraries. Please save your work in a Jupyter notebook and email it to us for review.
Data file: cruise_ship_info.csv (this file will be emailed to you)
Objective: Build a regressor that recommends the “crew” size for potential ship buyers.
Please do the following steps (hint: use numpy, scipy, pandas, sklearn and matplotlib)
1. Read the file and display columns.
2. Calculate basic statistics of the data (count, mean, std, etc) and examine data and state your observations.
3. Select columns that will be probably important to predict “crew” size.
4. If you removed columns explain why you removed those.
5. Use one-hot encoding for categorical features.
6. Create training and testing sets (use 60% of the data for the training and reminder for testing).
7. Build a machine learning model to predict the ‘crew’ size.
8. Calculate the Pearson correlation coefficient for the training set and testing data sets.
9. Describe hyper-parameters in your model and how you would change them to improve the performance of the model.
10. What is regularization? What is the regularization parameter in your model?
Plot regularization parameter value vs Pearson correlation for the test and training sets, and see whether your model has a bias problem or variance problem.
Most often, they’ll give you about 4 to 7 days to complete the project. Sometimes they may specify if they want you to write a project report or simply turn in an R script or jupyter notebook file.
Lessons Learned: This is an excellent opportunity for you to showcase your ability to work on a data science project. You need to demonstrate exceptional abilities here. For example, if you are asked to build a multi-regression model, make sure you can demonstrate a full understanding of the following advanced concepts:
(i) Feature standardization
(ii) Hyperparameter tuning
(iv) Techniques of dimensionality reduction such as PCA (principal component analysis) and Lasso regression
(v) Generalization error
(vi) Uncertainty quantification
(vii) Demonstrate the ability to use advanced data science techniques such as sklearn’s pipeline tool for model building.
(viii) Be able to interpret your model in terms of real-life applications.
Step 5: Skype Interview with Director of Data Science
After successfully completing the take-home data science challenge problem, I had to move on to the next step. This is now the 4th week of the interview process. I had a Skype interview with the director of data science. The entire interview lasted for about 30 minutes. At this stage, the main goal was to determine my level of enthusiasm and passion and to figure out if the current role would be a good fit. It was basically a general conversation about my background and my past experiences, as well as previously completed data science projects. The interviewer would like to determine if you are really passionate about the role and your level of motivation.
Lessons Learned: As already mentioned before, advanced preparation is key at this stage. Do your homework, be knowledgeable, and always ask questions to dig deeper into what the company is all about, the culture, strategic plans, visions, compensation, benefits, etc. Also, it’s always recommended to dress professionally and be in a location with no distractions.
Step 6: Onsite Visit
At this stage, if the company decided that they would like to move on with you, they will fly you out to where they are located. This would normally take place about 1 to 2 weeks after the Skype interview is completed. This is typically a full day of the interview, with several rounds of interviews with employees, directors, etc. Just as for the Skype interview, it’s important you show up in time, dress professionally, and be enthusiastic about the position. During the interview process, make sure to ask as many questions as you can about the company and your job role.
Step 7: Job Offer/Rejection
After the onsite interview, an offer is made typically within a week or two. They will typically notify you by phone. If it is an offer, they will provide instructions on the latest date when you have to get back to them with your decision as to whether you’ve accepted or rejected the offer.
In general, I’ve discussed the processes involved in a data science project. Any aspiring data scientist should be well prepared to face the lengthy and sometimes tedious interview process. The process could be long and tiring, but you will learn something from each interview process. I hope the guidelines provided here will be beneficial for anyone aspiring to become a data scientist.