Top interview questions data science job seekers fail

Are you a data science enthusiast looking for your first job in the field? Have you spent countless hours studying and practicing but still had no luck? What could be the reason behind this? It’s time to take a closer look at why you’re not getting that data science job, focusing on typical interview questions data science applicants encounter.

In this post, I discuss five reasons you may fail to get hired as a data scientist. From lack of experience or knowledge to inadequate communication skills, I will help you identify what is holding you back so that you can make changes and succeed in your career goals.

Read on to discover the top five reasons you are not getting that data science job.

What is the typical process for getting a data science job

Whether you are a recent graduate or someone with experience, the job search often starts with you reaching out to someone about a job posting and sharing a resume. It sounds simple, but I see people making mistakes here pretty often.

You’ll get to the interview stage if you do the first stages right. You would think that, on paper, very qualified candidates tend to do well in interviews, but this is not necessarily true. I believe that part of this has to do not with lack of knowledge but with lack of training in conveying one’s knowledge. I have interviewed academics looking to switch to the private sector who are clearly more than qualified from a technical standpoint but still don’t pass the interview.

Continue reading to find out some of people’s data science job search and data science interview mistakes. If you’re a new graduate looking for your first job, make sure to read my guide on getting your first data science job as well.

Top five reasons for not getting a data science job despite applying and interviewing

1. Apply for the right job

I can’t count the number of times I share a job posting on LinkedIn that says something like “Ph.D. in economics or related discipline required,” only to get 15 LinkedIn messages from Master’s students. If it says Ph.D. required, that means Ph.D. required.

You don’t have to hit all the requirements on a job posting to apply, but there are certain must-haves you need to identify. By reaching out about a job listing that you are clearly not qualified for, you are doing yourself a disservice because it shows that you are not capable of or not willing to understand the job description.

In general, it is a good idea to reach out to people who work at the company you’re interested in and ask them for a referral. Most companies pay referral bonuses, so people are incentivized to refer qualified candidates to get a bonus if the person gets hired. If you reach out to someone about a job you are not fulfilling the requirements for, you are just doing bad marketing for yourself.

data science interview questions - solving the puzzle

2. Be confident, not arrogant

There is a fine line between confidence and arrogance. I have been in situations where we have two candidates, one of whom seems slightly more skilled or knowledgeable, but we choose the other candidate based on other characteristics.

Remember, you are rarely hired to sit in a silo and produce output independently; in almost all jobs, you are expected to work as part of a team, and people want to work with nice people. This is more important than working with someone who knows everything. So, even if you are an expert on the topic and you can ace all interview questions, make sure to do so in a way that conveys that you are open to collaboration and constructive feedback.

3. Be able to explain why you want the job

Even if the company is not your top choice, if you want to get a job offer, you need to make a convincing case that you are interested in getting the job. In general, it’s hard enough to hire qualified people, and extending an offer to a candidate who does not accept is a waste of time and resources on the company’s part.

Hopefully, you are applying for jobs that you are excited about. For each interview, identify at least a few aspects of the job you are genuinely interested in. Those could be things related to your experience and expertise or areas that are new to you and that you are interested in learning more about. If it’s the latter, try to be more specific than just saying, “I know nothing about X, but I’ve always wanted to learn about it”. You should be able to articulate what it is about X that interests you.

4. How to prepare for data science interview questions: know how to sell your skills

The interview is a sales pitch for yourself. Even if you are a perfectly qualified candidate, the interviewers won’t know unless you can convey this. A lot of this boils down to knowing how to interview.

Data science interview questions are often vague on purse. Hence, when answering a question, it’s usually good to start by taking a step back to assess the problem and ensure you understood it correctly. At this point, you can recap the question and ask clarifying questions.

A mistake candidates often make is that they dive right into the solution and try to be very specific, but they may be completely missing the point because they misunderstood the question or problem in the first place. This is a sign that you can’t trust the candidate to figure out a non-trivial business problem independently.

By taking some time to make sure you are answering the right question, you are not only allowing yourself to come up with the correct solution, but you are also showing that you are a person who can be trusted to work independently on an ambiguous problem. It is also good to reason out loud when thinking about a solution.

5. How to answer data science interview questions: don’t display red flag warnings

Hiring someone is always a risky investment. You spend time and resources to collect as much information as you reasonably can, then decide whether to hire them. Even if the candidate is qualified and everything else seems good, some red flags can dissuade a recruiting team from hiring even the apparently most qualified candidate.

data science interview questions - red flags

Be able to explain your own work

One scenario that happens more often than you would think is a recent graduate who cannot explain their thesis! You should be able to explain your thesis at a high level (think, to a non-technical audience) and go into the technical details. If you can’t do the former, this shows that you can’t be put in front of non-technical stakeholders without some serious coaching. If you’re not able to do the latter, this raises serious questions about whether you actually wrote your thesis yourself. You are not expected to know everything about everything, but you need to be able to show that not only do you understand your own work, but you can also explain it to others.

Don’t give vague and fluffy answers

Another big red flag is vague answers to technical data science interview questions. Avoid being vague! I can’t count the number of times someone says, “I’ll run a regression of customer acquisition on X features,” but when I dig deeper, they cannot specify what they mean by customer acquisition or exactly what features they would use and why. If I’m interviewing someone for a skilled technical position, I expect a level of precision along the lines of “I will use a panel data set where the cross-sectional dimension represents customers and the time dimension represents months. My dependent variable measures customer spending of customer i in month t, and I regress it on a customer fixed effect, lagged customer spending, and other customer and/or time-varying features”.

There is nothing wrong with sometimes saying, “For this particular problem, there is a very specific technique called X that I think would be applicable, but I’m not an expert in this area, and I’d need to read up on the literature first”. However, if you are providing a solution, it is unacceptable to be vague and fluffy.

data science interview questions - vague and fluffy

Avoid negativity

Voice your past jobs in a positive light. Everyone understands that if you’re interviewing, you’re looking for a job that’s better than your current one in some regard. However, an optimal way of expressing this exists. For example, you can say, “I am looking forward to advancing my skills by learning new things,” instead of “My current boss is an idiot.” Additionally, if you have had this type of experience in all of your past jobs, people may suspect that you are the problem.

Be reasonably honest about why you want to change jobs; this question comes up often, and answer the question professionally and elegantly.

Takeaways

The job search is a matching and sales problem. You need to find a job that matches your skills and interests, and then you need to convince the employer that this is the case. The flip side is also true; unless you are desperate and ready to take any job, the prospective employer has to sell you on the opportunity.

Focus on developing the right skills for the job, be smart about what jobs you apply for, and learn to interview. It helps to learn to answer questions using the STAR interview method. Finally, don’t forget to be nice and show genuine interest. This goes a long way.

Still have questions? Feel free to reach out!