You’ve done the work. You’ve built, designed and studied. You’ve navigated the ins and outs of Javascript, Python and SQL. Looking ahead, you see your next hurdle: landing the job.
Keeping your head above water during your first round of interviews for a data science role might feel like an impossible task, but take it from someone who’s been through it already: You don’t need to be an expert to be the ideal candidate.
That was certainly the case for Alex Rich. Coming from an academic background, he was out of his comfort zone during the product management portion of his first interview. “There’s no ‘product’ in academia,” he explained. But Rich didn’t let that slow him down during the process. Instead, he focused on the gaps in his skill set and exhibited them as growth opportunities — for himself and for his potential as a new hire at Flatiron Health.
With the interview process behind him, Rich shared his perspective on what to expect as a potential candidate and how to prepare for those first data science interviews. He also offered valuable advice on what technical skills to prioritize (think: getting comfortable with SQL). If his experience is anything to go by, well-thought-out stories behind past projects and experiences are key. Remember, you did the work to get to where you are — you just need to frame it like a data scientist would.
Tell us about your first interview for a data science role. What was the format of the interview, and what’s a key lesson you learned from that experience?
After finishing graduate school, I looked for healthtech data science jobs in New York. My first experience interviewing for a data science role was with Flatiron — I was excited and nervous.
The first interview of the day focused on product management skills, which was particularly nerve-wracking because I had no clear idea what that meant. The interview format was a case study; the interviewer described possible features for a hypothetical new product. I worked with him to prioritize which ones to build first, and explained how I’d test our hypotheses about what features mattered most. I felt pretty out of my depth in this interview, but ultimately landed the job.
What I learned from this experience was that you don’t need to be an expert in everything to land a job. It’s OK to have growth areas; if you can show a willingness to learn and be coached, and an ability to think through new and unfamiliar types of problems, you can still be an outstanding candidate.
What is the most important thing you do to prepare for a data science interview?
When talking to candidates coming from a more academic, machine learning or statistics-focused background, one thing I emphasize on the technical side is not overlooking familiarity with SQL. While it’s not always the most exciting tool, it’s often the way data is stored, manipulated and moved around, so knowledge of how to use it is essential for a data scientist to work effectively. Because of this, it’s a frequently tested skill in data science interviews, particularly in early rounds.
Think through not only what skills you showcased in your past projects, but also why those projects mattered and what important problems they solved.”
What advice do you have for someone preparing for a data science interview at your company?
Flatiron splits most of the data science interviews into two parts. In the first part, candidates are prompted to walk the interviewer through a previous project, emphasizing product or machine learning skills. In the second part, they work through a predefined interview problem, such as a case study or pair coding exercise.
Because many of our interviews start with this open-ended component, having stories from your past experiences at the ready is an important part of preparing. And at Flatiron, we’re very focused on impact — one of our company values is “start with why” — so be sure to think through not only what skills you showcased in your past projects, but also why those projects mattered and what important problems they solved.
