Don’t Miss this Comprehensive 7 Step Process to Ace Data Science Interviews!

Pranav Dar 19 Jul, 2020 • 11 min read


Are you looking to get a break in data science but struggling to clear interviews? Are you scared of getting into data science interviews? Or you just aren’t sure what to expect in these interviews?

You might even know the data science tools and techniques relevant to the job. And yet the employers keep rejecting you. It certainly doesn’t help when job descriptions require multiple years of experience for a seemingly entry-level role!

I’ve been there. I have a rich background in learning and development, a non-technical and non-data science field. It took me months of hard work and disciplined effort, and numerous botched interviews, before finally landing a data science role here at Analytics Vidhya.

My aim in this article is to list down a framework we have come up with to help you ace your data science interviews. We have drawn on our own experience plus reached out to multiple data science experts to craft this comprehensive 7 step framework. It has already helped multiple people, including me, streamline their data science interview preparation.

This framework is part of the ‘Ace Data Science Interviews‘ course. This course has been created based on hundreds of interviews we have taken and worked with companies to help them find data science talent.


Here’s the 7 Step Data Science Interview Process

The 7 step process starts right from the stage you start researching the different roles that interest you. And it goes all the way up to the completion of in-person interviews.

Keep in mind that this is a comprehensive framework. You might not have to go through each and every step in your interview journey.

  1. Understand the Different Roles, Skills and Interviews
  2. Getting Ready for Interviews – Build your Digital Presence
  3. Prepare your Resume and Start Applying!
  4. Telephonic Screening
  5. Getting through the Assignments
  6. In-Person Interaction(s)
  7. Post-Interview Steps

In this article, we will analyze all these steps and look at the mistakes people make in each step. Additionally, all these 7 steps are packaged together neatly in a wonderfully illustrated infographic at the end of this article. Keep it handy throughout your interview preparation!


1. Understand the Different Roles, Skills and Interviews in Data Science

Have you picked the role you want to pursue in data science? The most common answer I get from most folks is “I want to be a data scientist”. What else is there in data science?

The first thing you need to understand is that there are a variety of roles in the data science ecosystem. A typical data science project has a lifecycle that’s made up of several functions. A data scientist is just one component in a successful data science project. Here’s a quick run through of the different job roles that currently exist:

  • Data Engineer
  • Data Scientist
  • Business Analyst
  • Data Analyst
  • Data Visualizer (the Storyteller)
  • Statistician
  • Data Science Manager
  • Data Architect
  • Machine Learning Engineer
  • Computer Vision Engineer

I could go on but you get the idea. I recommend spending time to research these different roles and pinpointing one that you see yourself in. We have crafted an intuitive question sheet as part of the ‘Ace Data Science Interviews‘ course that helps you make this decision.

The next step is to understand the skills required in these roles. For example, you need to have strong Python and Software Engineering background for a data engineer role – but communication skills are not that critical. On the other hand, if you want to get into a Business Analyst role – you need to have good communication and problem solving skills. You may not need to know Python.

Now that you know which role you want to pursue and what skills are required to get there, it’s essential to map the kind of interviews you might face. Again, these interviews will differ from role to role

A data scientist will be grilled on his/her machine learning knowledge, grasp on the tool(s), domain expertise, communication skills, among other things. A data engineer will primarily be tested on his/her software engineering and programming skills. Understand the different nuances expected in your role and prepare accordingly.


Common Mistake People Make at this Stage

I’ll circle back to my original question – which data science role do you see yourself in? Most people I’ve seen make a huge mistake at this stage. They don’t put in the time to understand the nuances of these different roles.

Hence, their interview preparation remains the same regardless of the role. Don’t make this mistake! Even if the interview format is the same, the expectation of the interviewer would change depending on the role you choose.

My suggestion would be to talk to people in your network who are in this field. Pick their brain on what their understanding is of each role. You can also leave your questions for me below this article – I will be happy to answer any queries.


2. Getting Ready for Interviews – Build your Digital Presence

Done your due diligence so far? Good, because it’s time to move on to step #2 – building your digital presence.

More than 80% recruiters we spoke to admitted they check a candidate’s LinkedIn profile before calling them for an interview. That’s right – we are living in the middle of a digital revolution. Simply relying on a 1 or 2 page resume is not enough. The recruiting firm wants evidence to back up the claims in your resume.

The good news – there’s no dearth of options to do this:

  • You should have a LinkedIn profile. It should be updated and optimized according to the role(s) you’re applying for. Applying for a data scientist role while showcasing a non-technical background will not give the correct impression
  • Create a GitHub account. Programming is a vital cog in the data science machine. Uploading your code and projects to GitHub helps the recruiter see your work first-hand. Nothing more convincing than a well-documented code!
  • Regularly answer data science relevant questions on Quora. This conveys your understanding of the subject matter
  • Start publishing your learning in blog form. Learned something new? Write about it. Put it out in the open. Ask the community for their feedback. That’s how you build credibility and enhance your chances of getting an interview
  • Apply to speak at meet-ups and conferences. Build your reputation in the data science community by attending events like DataHack Summit. If you can get through as a speaker, the job offers might just start rolling in

The options are endless! You need to pick the medium(s) depending on the roles you want to apply for and your own strengths (for instance, a Business Analyst might not need a GitHub account).


Common Mistake People Make at this Stage

Don’t leave this step till the end! I’ve seen people creating a GitHub profile a few hours before applying to a job. By the time you are doing or applying for interviews – you won’t have the desired digital clout. This last minute dash simply doesn’t leave enough time to optimize your details according to the role requirements.

The same story applies to LinkedIn. It takes patience and commitment to build your network. It took me over a year to connect with the right people in data science and build up my profile there.

Our advice – you need to do this at least 6 months before you apply for the jobs.  This will not only ensure that you get enough time to build your presence, but also take away the pressure from the overall process.


3. Prepare your Resume and Start Applying

If I had to pick the toughest step in the data science process, this would be it. Yes, I am putting this above the in-person technical interviews. You can study and brush up your data science knowledge. But crafting the perfect resume? That’s a different ball game altogether.

I’m sure a lot of you would agree. Every recruiter and hiring manager has their own criteria for judging candidates. So designing a crisp and concise resume is the first thing you should consider. A few tips to do this:

  • Make sure your resume reflects the relevant technical skills you’ll need for the job. You might be a PowerPoint whizz, but that should not be your key skill for a data engineering role
  • Create different resumes for different roles. Talk to the HR or an alumni of the company and see what the expectations are. A startup will usually have very different hands-on expectations as compared to an established corporate firm

Once your resume is ready, expand your job search. Remember that job portals are not the only way to apply for data science roles! In fact, they are the least effective manner of searching for jobs. There are more than 9 different ways to apply for jobs and job portals happen to be just one of them.

We discuss each of these 9 ways in the ‘Ace Data Science Interviews‘ course. This includes tips and tricks to optimize your applications in order to land that dream interview!


Common Mistake People Make at this Stage

This one is easy – the most common mistake I see people making at this stage is they go about doing passive job searching. They do all the hard work, spend days and weeks learning the right things, and then flush it away by being unaware of how to apply.

Image result for frustrated

Most of us typically come back from office, and then log into a few popular job portals and apply there. It’s pretty easy, after all. Your resume is already uploaded – all you do is find a relevant job and click a button. This strategy no longer works. The recruiter will have hundreds of applications on the portal, all looking almost the same.

Impossible to distinguish between that. It becomes a game of chance. Even if this strategy works, it will be months before you get any thing meaningful from this channel.

We are in the digital era, folks! Get creative! You spent all that time building your network in step #2, put that to good use. Most data science interviews you land will be through non-job portal channels.


4. Telephonic Screening

If you have reached this stage – congratulate yourself! It is time for your first real interaction with the recruiter.

Depending on the company, there could be a call only with a recruiter, or only with the hiring manager (or both). But the fundamentals remain the same. If the hiring manager is taking the call – you should expect a few technical questions as part of the process.

Since this is a telephonic or a virtual call setup – you can keep your answers to most of the common questions ready. I wouldn’t recommend reading them straight from the screen, though. You should be prepared with the broad pointers and take it from there.

Take this round as seriously as all the other steps. A casual vibe is enough to throw the recruiter off. Additionally, always minimize distractions in the room where you’ll take the call. You can also take notes throughout the call for reference later. I have givena  comprehensive run down of the do’s and dont’s to ace the phone interviews in the course.


Common Mistake People Make at this Stage

Too many people fail to ask questions at this stage. This is a great opportunity to understand more about the role, the company, and your fit within that ecosystem. Don’t just stick with the old “what are the work timings” questions! Showcase your curiosity and passion for the role.

You should be prepared to ask questions which you can not find in open public information. These are the small things that add up big time in the overall scheme of things. The hiring manager will appreciate your interest. That’s how rapport building is done.


5. Getting through the Assignments

If the telephonic round went well, there’s a good chance you might be asked to do an assignment. Now, not every company has this round. It varies from role to role and project to project. But it’s best to be prepared, right?

You can expect to face one of the below types of assignments:

  • Take-home: You will typically be provided with a problem statement and a dataset and asked to solve the challenge. Take-home assignments can have a deadline of a few days
  • On-site: This is often integrated into the in-person interview rounds. You might be asked to spend anywhere between 3-8 hours on this

Typically, these assignments act as filters and would usually be basic in nature to ensure you have the skills you claim on your CV. The submission could be in the form of a presentation, Jupyter notebook or a submission on a self-evaluation platform.


Common Mistake People Make at this Stage

Aspiring data science professionals tend to only meet the bare expectations in the assignment. Asked to build a model? Sure, here’s a report about the results. The thought of going beyond that, perhaps exploring the data and discerning patterns, doesn’t occur to most candidates.

Whatever the assignment – you should go the extra mile to solve the problem. Don’t stop by just doing what is expected. This is your chance to impress the hiring manager even before the face-to-face interview starts.

These assignments will map closely with the role you’ve applied for. A wonderful opportunity to explore what you’ll be potentially getting into. And if your telephonic round didn’t go very well, this offers you a chance to get back on track.


6. In-Person Interaction(s)

You’ve researched the role, crafted a crisp and exquisite resume, successfully applied and cleared the telephonic and assignment round. You know what’s coming next – it’s finally time for the face-to-face interviews!

You can safely expect to face multiple rounds and formats of interactions. You’ll meet with plenty of people throughout these rounds. In-person interviews can take anywhere between half a day and an entire day to finish. The different people you might interact with:

  • Hiring manager
  • The data science project team
  • Project manager/leader
  • Other stakeholders
  • HR person

A hands-on hiring manager could also ask you to sit with the team through the day and brainstorm on a problem. The assignment round we saw earlier could also factor into this stage.

You’ll be judged on your structured thinking, analytical and logical reasoning, puzzle solving skills, programming knowledge, machine learning techniques, among other things. The use of whiteboards has become quite common now in data science interviews. Writing down SQL queries, or explaining your thought process – be ready for all these aspects.

Check out this comprehensive interview guide to prepare for variety of interview rounds and questions


Common Mistake People Make at this Stage

One pet peeve of mine is when the candidate doesn’t ask meaningful questions. I know we covered this in the telephonic interview but it’s even more essential in a face-to-face setting. Asking questions is a sign of curiosity and interest in the role. If you have already asked the questions about the role in a previous round – you can always ask the interviewer about their journey in the organisation.

The in-person rounds can be quite grueling. There are just so many of them! Candidates at times don’t expect this and start losing their focus halfway through the day. Mental preparation is as important as the rest. Patience is a virtue, and a data science professional is expected to posses bucket loads of it.


7. Post-Interview Steps

Finally, once you have gone undergone the in-person interviews, you should follow-up with a thank you note. Also, make sure you’re honoring any commitment you might have made, such as sharing a past presentation or a piece of code). You’ve done the hard work, it’s now time to wrap up things and bring that dream role home!


Common Mistake People Make at this Stage

Regardless of how you felt the interview went, you should not be unprofessional at this stage. Yes, even if you don’t hear back from the interviewer after you sent a thank you note.

This can come across as desperate. The data science community is still a small world and your reputation will precede you. Don’t do anything to harm your chances of getting another interview opportunity in the future.


End Notes and Infographic

And that wraps up the 7 step framework for data science interviews! Here are a few resources you should go through to boost your chances of acing the next data science interview you’ll face:

And as promised, here is the infographic we have created on this 7-step framework. Download it, save it, and use it as a checklist for your next data science interview preparation!

Pranav Dar 19 Jul 2020

Senior Editor at Analytics Vidhya. Data visualization practitioner who loves reading and delving deeper into the data science and machine learning arts. Always looking for new ways to improve processes using ML and AI.

Frequently Asked Questions

Lorem ipsum dolor sit amet, consectetur adipiscing elit,

Responses From Readers


Shantanu 21 Feb, 2019

Hey Pranav, Very well put up article. Thanks for sharing, some insights were really good and set a new point of view for me!

42 Must Know Data Science Interview Questions and Answers - DatabaseTown
42 Must Know Data Science Interview Questions and Answers - DatabaseTown 14 Mar, 2023

[…] This article will also be helpful for you in interview preparation. 7 Step Process to Ace Data Science Interviews. […]