It’s tempting to take a combative approach to job interviews.
Applicants enter the hiring process on the outside looking in, desperate to breach the castle walls and join their inquisitors inside Camelot.
Such encounters, however, are not obstacles to be overcome, but an exercise in allyship. As Jaymee Sheng, a machine learning engineer at the Healthtech company Ginger, explained, “Think of your interviewers as collaborators, which they will be if you end up working there, rather than adversaries who are trying to ‘get you.’”
Interviews, after all, are conducted with the intention of welcoming someone into an organization; hiring managers ask questions to gain insight into a candidate’s accomplishments — not in an attempt to undermine their chances at landing a job.
“One lesson I learned is to prepare as much as you can for any areas you are less familiar with, and to be honest when you don’t know something,” Sheng said. “Just relax and act like you are with a colleague.”
It isn’t just the soft skills that interviewers are looking for, Sheng added. Expect a technical component that, if nothing else, tests your knowledge and demonstrates whether your skill set matches the requirements of the job.
To learn more about data science interview dynamics, Built In San Francisco checked in with two local experts to ask them what they look for when they’re asking questions, and how prospective employees can prepare themselves to join their team.
Tell us a little bit about your first experience interviewing for a data science role.
When I decided to move to the tech industry from academia, I was very nervous! The interview lasted a half-day and included a series of 1:1 meetings with several other data scientists and managers to review my experience in research and consulting. It’s always fun to talk shop about your research and learn what interesting things other folks are up to. There was a white-boarding section where I was asked to write code as part of a conversation. The white board part was a bit adversarial, and the room was very small. There was a take-home coding assignment as well. The interview was hard, but it really gave me an opportunity to reflect on the experience. I went into my next interview more articulate about my strengths and more clear and confident about the kind of opportunities I was looking for.
What is the most important thing you do to prepare for a data science interview?
My top priority is to review the job posting and look up the company. I want to know more about what the position is, what the company does and how they do it. When I review the position, I have an eye for where I have matching experience and what parts of the job description might indicate interesting growth opportunities. If I’m working with a recruiter, then I take some time to chat with them for any relevant information they have about the business, position and company culture. I always cross-check LinkedIn to see if I know anybody at the company.
For in-person interviews, I bring extra hard copies of resumes, a notebook to jot down notes and a pen that I know works. If there are hints at company culture, then I may adjust my interview outfit to more or less formal as well. I try to make sure that I get a good night's sleep, a good breakfast and get there early so that I’m not hurried.
I’m way more interested in a messy smaller project that you did yourself than a perfect Kaggle project where you followed a recipe on the web.”
What advice do you have for someone preparing for a data science interview at your company?
The first priority is to review how you present yourself. Make sure that your LinkedIn profile is professional and up to date. Review your resume for content, organization and clarity. Highlight some interesting projects that you have done in the past with a short statement of purpose and relevant skills.
I’m generally looking at the project process more than the algorithms: How did you get the data? What kind of work did it take to organize that data? What did you find in your data exploration? What choices did you make, and why? Nobody will know all the algorithms, but if you can explain why you picked certain algorithms, then I get a better idea of your building process. For entry-level folks, I’m way more interested in a messy smaller project that you did yourself than a perfect Kaggle project where you followed a recipe on the web.
During the interview, I’m looking for people to be curious and engaged. My team is building things together, so communication and collaboration are key skills. As data scientists, we must be able to communicate with all kinds of stakeholders. We need to be patient and methodical while explaining concepts, as well as flexible on approaching a concept.
Tell us about your first experience interviewing for a data science role.
Believe it or not, my first data science interview was actually for my current role at Ginger (now Headspace Health). After a preliminary 30-minute phone interview, I was given a week to complete a take-home project, which should only take a few hours and is intended to mirror the type of work machine-learning engineers at Ginger do. After that, I was invited to a virtual on-sight meeting, during which I presented my project, had a social lunch hour with the entire team and met one-on-one with several members of the data science team, as well as our product partners.
What is the most important thing you do to prepare for a data science interview?
I always review the projects I have done and make sure I know them well enough to comfortably talk about any particular part of the project in depth, whether it’s an implementation detail or the rationale behind a methodology. I also think about how my learnings from those projects can translate to the position I’m interviewing for. Most companies are much more interested in how you arrived at your results than the results themselves.
Most companies are much more interested in how you arrived at your results than the results themselves.”
What advice do you have for someone preparing for a data science interview at your company?
Be yourself, know why you want to work here and know what you can bring to the table. Ginger is full of passionate, mission-oriented people, so it’s important to demonstrate that you care about the why behind your work. Curiosity and desire to learn is key, as data science as a field is evolving rapidly and our company is growing quickly to deliver quality mental healthcare to an ever-increasing population.
