Prepping for The Data ScienceTechnical Interview
As a recent graduate of an online Data Science bootcamp, I have been getting ready for the job search and a large part of this preparation has been preparing for the dreaded technical interview. Thankfully, the program that I graduated from included a mock interview at Skilled, which I cannot endorse enough. Having a practice interview under my belt, I feel much more comfortable moving forward, and I encourage anybody worried about a technical interview to check them out.
Even though my interview was for practice, I took the preparation seriously. As I was looking into what a technical interview could entail, I found myself getting more and more nervous. Every article I read contained at least one more thing that I hadn’t prepared. As I discovered more areas to study, my anxiety continued to climb. When the day for my interview came, I felt like there was no chance I was going to pass. However, when the dust settled, I passed with a score of 4 out of 5. The interview went very well and I had a really pleasant conversation with my interviewer. Having gone through this ordeal, I decided to share some advice and hopefully save someone from the anxiety that I experienced.
Recognize Your Weak Points:
If you have a limited amount of time to study for an interview, you really don’t want to waste any of it working in areas where you’re already strong. Identify the areas where you feel less confident, and devote more time to strengthening those skills. For me, this was python algorithms. I was terrified of the coding part of the interview and couldn’t shake the image of sitting at the computer staring at a problem that I had no idea how to solve. I spent a good portion of my time preparing for my interview by practicing algorithm problems on hackerrank, Project Euler and Interview Cake. This practice paid off as I was able to sail through the python coding portion of the interview. I was extremely glad that I had spent the time studying coding rather than areas where I felt confident.
Prepare for all Parts of the Interview:
While it is important to recognize where you can improve, it is also important to remember that there is more to the interview than just coding. Aside from programming, you will also need to know statistics, probability, experimental design, and modeling practices. Quiz yourself in these subjects and see if you will need to spend time reviewing them. It is easy to overlook some of these areas because they are the bread and butter of a data scientist, but it is worth it to make sure you know the theory behind your models. For my interview, I was blinded by my algorithms preparation, and neglected to prepare my SQL coding. I felt like I was comfortable with SQL and consequently hadn’t reviewed it. This showed in my interview and my code was a bit sloppy. Thankfully I had a well rounded preparation aside from this, so the rest of the interview went incredibly smoothly.
Practice:
This is by far the most important thing you can do to prepare for the interview. Practice your python, practice your SQL, practice answering theory questions. Whatever weak point you have identified, do as much as you can to increase your skills. Below is a list of resources to help practice for a technical interview. I encourage you to browse through them and see if they are helpful.
In addition to practicing individual skills, there are several places where you can do an entire mock interview. I went through Skilled for my interview, but there are other sites like Pramp that will also help you practice actually interviewing. This experience is very valuable because it lets you get used to the environment of interviewing. Doing coding problems on your personal computer isn’t too tough, but solving them in front of somebody who is analyzing your methods is very stressful. The more you can get used to this type of pressure, the more comfortable you will be in an actual interview.
Talk to Your Interviewer:
This was advice that I read repeatedly, but didn’t fully understand before the interview itself. your interviewer wants to see how you think and problem solve on your own, and as a team. If you don’t know the answer to a question, say that. Tell them what you do know that’s related and talk through how you would try to answer the original question. If you are given a coding problem that you don’t know how to solve, talk through the problem and what your code needs to do. If you’re stuck on how to resolve a step, your interviewer will likely work with you to come to a solution. My interviewer helped me walk through a whiteboard question because I told them what I wanted from my code. As I was identifying what variables I was going to need, they were able to offer a few suggestions to make my pseudocode more efficient. We were able to collaborate together to solve the problem quickly. By talking to your interviewer about your process, you show that you can work cohesively with a team and what it is like to work with you as a programmer.
Relax:
The most important thing about interview preparation is to try and stay calm. Studying and practice will be more effective if you are relaxed. The interviewing process is naturally stressful, and a technical interview is exceptionally so, but do try to remember that each interview is a learning experience. If an interview does go poorly, then you will know some areas where you can improve for next time.
Below is a list of resources that you may want to check out if you’re having trouble preparing for an interview. Hopefully they will be as helpful with your preparation as they were with mine. Thanks for reading!
Resources:
Useful Guides:
https://learntocodewith.me/posts/technical-interview/
https://towardsdatascience.com/preparation-guide-for-data-science-interviews-69932d30c7c4
Practice Interview Services:
Practice Coding Resources:
Statistics and Modeling:
https://medium.com/data-science-insider/160-data-science-interview-questions-14dbd8bf0a08
