Data science has a steep learning curve, but there’s a popular misconception that there’s a specific educational background or “secret sauce” required to enter this field. In reality, data scientists come from a range of backgrounds—some of them are career switchers who bring knowledge from other industries or fields such as software engineering, statistics, or mathematics. (Take Springboard alum Adrien Saremi, who used his physics degree to gain a competitive advantage over other students in Springboard’s data science bootcamp.)

At Springboard’s annual Rise 2020 virtual conference, Ravi Ram, director of sales and admissions at Springboard, interviewed three data scientists from Google, Spotify, and The Athletic on how they ended up in data science after switching from different industries, and what advice they have for aspiring data scientists who want to find their niche.

Here’s an excerpt from their conversation, which has been lightly edited and condensed for clarity. 


  • Sarah Hooker, research scholar at Google
  • Chloe Liu, senior director of analytics and data science at The Athletic
  • Pasquale Prosperati, data scientist at Spotify

How did you end up in data science? What made you interested in the field?

Chloe: I started as an electrical engineer on the hardware side, building circuit designs in college. Later on, when I studied statistics, I found that my passion was data. There’s a lot of detective work involved and everything you find out from the data tells a story and gives you a new perspective on things. 

When I started, there was no such thing as a data scientist job title, there was only a statistician. Usually, statisticians would work in finance or as actuaries. So I started in the finance industry building financial models and then moved into e-commerce, using statistical models to decide what the pricing should be for hotel rooms. 

I taught myself data science programming languages and algorithms to where I am today, where I lead a team of data scientists and analysts. 

Pasquale: When I started my career, data was mostly just a side project. After getting my master’s degree (in Mass Media and Communication) I started working at a market research company responsible for selling data, where data was just a [commodity]. Then my wife got a job in New York and we moved there and I had to decide what to do with my career. I started learning more about how to make data a part of my job, so I taught myself SQL and tableau from scratch. 

Sarah: Right now I’m doing peer research in the data science field with the goal of open-sourcing that knowledge. I got there by learning data science and then moving into engineering and doing machine learning recommendations and algorithms and then teaching machine learning. 

What do you enjoy the most about working as a data scientist?

Chloe: The thing I enjoy the most is changing people’s perspectives: When I find something that nobody anticipated and change the entire course of a decision. 

Pasquale: Getting a question from someone and being able to answer it with data and having all these tools at your disposal to do so. At the same time, that’s also the scary part. There’s always more that you don’t know than what you do know, so sometimes it’s a bit overwhelming. 

What was the hardest part of your journey to becoming a data scientist—especially if you came from a nontraditional background?

Sarah: My pursuit was very lonely at times. It’s hard when you’re pursuing something in a very solitary way where it’s just driven by your sheer interest and passion. I think that’s part of why it got so much easier when I started joining organizations where my colleagues were doing the same thing.

Data science has such a steep learning curve because it combines statistics, data analytics, machine learning, and even some software engineering. Many practitioners talk about having impostor syndrome. Is this something you experience?

Pasquale: 100% on the impostor syndrome. I often feel like, what am I doing here? The thing I always try to do is switch it around. You have to not let it paralyze you. Focus on one step at a time rather than the end goal. Every complicated machine learning problem starts with something small. Decide what you bring to the table and where you can add your skills one at a time to explore this bigger world. 

Chloe: Everybody has impostor syndrome. Prior to my current job I worked with a lot of startup founders to establish their data efforts in the company. A lot of these founders had impostor syndrome themselves. They’d be like, okay, am I good enough to be a CEO? It’s the same problem for everyone and it’s all about mindset and perspective. 

Sarah: I think impostor syndrome is amplified for people who come from nontraditional backgrounds or who don’t rely on typical signals. I’m kind of [an anomaly] at Google Brain. Most of my colleagues have PhDs, some have multiple PhDs. And in fact, what I found is that impostor syndrome tends to hit you if you do a self-driven path because it’s tricky to measure progress. 

How do you make sure you’re always growing your skills and making progress in your field? 

Chloe: When it comes to math problems, I just avoid them. Just kidding. Knowing your strengths and weaknesses is very important. My weakness is math problems, so this week, I’m going to make a tiny bit of progress on improving how I read math problems. A lot of times, the problem is not as complicated as you think. There are many paths to Rome. Sometimes the easiest solution is the solution. 

Springboard Rise panel data science

What advice do you have for aspiring data scientists if they’re interviewing for jobs but they don’t have the ideal experience employers are looking for? How should they approach the job interview? 

Pasquale: Job interviews are like dating. It has to fit both ways; it makes no sense to try to fake your way into it. Personally, I always feel like being transparent is important: Knowing what you’re good at, what you bring to the table, and what you want to learn. Most jobs ask for everything, and you probably won’t be able to bring everything to the table right away. Be clear about what you can do now, what you’ve done in the past, and what you’d like to learn.

Chloe, you’ve been a hiring manager for several data science teams. What are you looking for in that first interview? What stands out on a resume? Do candidates actually need a master’s degree?

Chloe: The first answer is no, I never look at their degrees. Unless it’s R&D in the data science field—then yes, having a Ph.D. with research experience is very important. But I can tell you, for 80% of the jobs out there, you don’t need a Ph.D. I have three roles open on my team right now and I’m doing maybe 7-8 interviews every week. One of the things that we as hiring managers look for is technical skills: you know how to run SQL, you’ve built regression models and you’ve solved certain types of problems. 

The key thing for me when I’m looking for a junior candidate is their awareness. Do they know that they are not that great at SQL? Are they willing to admit it? To us, that shows your potential to grow. You can’t grow if you don’t realize that’s an area you need to improve. 

What are some ways you can demonstrate your willingness to learn and the soft skills hiring managers are looking for during the interview process?

Sarah: It depends on whether you’re interviewing at a small company or a company like Google where there’s much more of a rigid, well-established process. In the same way that we have databases for coding questions, I’ve noticed a tendency to ask certain questions to measure technical data science skills. Personally, I don’t agree with that as a way to measure a holistic view of aptitude, but that is the world we’re in. 

Show me an interesting problem that you’ve worked on and make it personal to you. I think we’re at the stage where there’s enough technical talent but I want to know why you first got involved in data and that can come through your choice of project. Hiring managers want to talk to someone who’s excited and who excites them. 

Hiring managers remember people and passion; they don’t remember skills and projects. What are some projects that you’ve been excited to work on?

Pasquale: Joining Spotify was really exciting to me. Everyone loves music the way I love music and everyone is a musician or has worked in the music industry. We worked on the fandom side of things to create connections between artists and their fans. I think one of the most beautiful things is going from sitting in a room with the data to creating events where you actually get to see those people and know that data-enabled this connection between artists and fans. 

Chloe: I mentor a lot of data analysts. I really want to teach analysts about everything other than their technical skills. The other side of the story is the human factor in your day-to-day job: how do I manage my stakeholders? How do I get to the deep root of a problem? When I’m coaching my team I think that’s one big thing that’s missing today: there’s no school for that. Everyone teaches you about SQL but to get to the next level is to answer the human factor in your day-to-day work. 

For more Rise 2020 coverage, check out posts on how data science can be leveraged for social good and tips on transforming your career in a post-pandemic world.