If you use Spotify to stream your music, you’re likely familiar with Spotify Wrapped, the personalized end-of-the-year data recap of everything you listened to — from the ska music you secretly love to the binaural beats playlist that gets you in the zone for coding. Even non-Spotify users have probably seen the splashy (and often self-deprecating) Wrapped summaries shared on your social media feeds.
Last year, Mindy Seto got to work as a Data Engineer on the team that put together Spotify Wrapped 2021. Mindy typically is a Senior Back-End Engineer at Spotify, but she jumped at the opportunity to temporarily sit on the data team to work on Wrapped, which is Spotify’s single biggest campaign of the year.
“Working in data, it was very fun to see the story and translate that into something where other users can understand,” says Mindy, who’s worked at Spotify since 2018. “Usually Java and back-end people don’t get to interact with the user-facing, UI, web kind of thing. But with data, I could say to my friends and family, Oh yeah, I know exactly how Wrapped came about.”
Learn something new for free
Mindy’s role as a Data Engineer on Wrapped was to build the data pipelines that then analyze listeners’ data using SQL and Scala. With 456 million Spotify users and terabytes worth of listening data to work through, Wrapped is a massive effort that takes from summer to the end of the year to complete. (You can watch Mindy and other Data Engineers on the Spotify Wrapped team talk in-depth about the whole process here.)
Here’s how Mindy got started in tech and learned the data skills necessary to work on the data team for Wrapped, plus her advice for folks who are interested in exploring different areas of tech.
What got me interested in the job
I did a traditional computer science background, and got a degree from Johns Hopkins University. After graduation, I had another software engineering job at a government defense contractor in Virginia.
I always wanted to work for a company whose product I know and interact with. I really use Spotify a ton and I love music. I also wanted to work for a company that has a really cool app. So when I was applying to companies during the job search, that’s what I thought about.”
How I got in the door
“It was very traditional: I went to the Spotify job listing page. The only listings that I saw were for Senior Engineers, and I only had 2 years of experience at the time. My background fit most of the criteria, other than the years of experience, so I just went ahead and applied anyway. By doing that, I still got reached out to by the recruiter [for the Senior Engineer position] and everything, and went through that whole interview process.
At Spotify, we have something called ‘embed opportunities,’ where someone will post a listing of what they need for a short period of time — usually the scope or a project is a few months. Any engineer can go and pick it up, and it’s kind of a way to do different Spotify projects in different parts of the company. What happened last year was, I saw a listing for Wrapped engineers to join, and I reached out. I was like, This is a great project I want to work on.
For Wrapped, I worked as a Data Engineer for about 6 months, because all of the back-end positions were filled up. I thought, I can try data engineering. At Spotify, it’s pretty cool because we have a bunch of different internal tech learning courses. I really wanted to learn more about data, so I took a course. A lot of it was learning on the job, but it was nice because I had a lot of resources and engineers helping me out, so it wasn’t too bad. Also, with my background in back-end engineering, it’s not too hard to pick up new languages.”
What I actually do every day
“So the Data Engineers took a Wrapped concept from SQL code, and had to translate it into the Scala pipeline. We were the ones that took the lead and actually made it into these giant datasets that power the back end.
Other people I worked with get together and come up with cool ideas for what they want to see in Wrapped, like the data story. For instance, last year we had a movie soundtrack thing, where it was like your music as if it were a movie. From that, the Data Engineers validate whether the data stories are something we can actually do, then produce the end-points as usual, and then mobile goes from there and works with design.
The biggest thing I learned was seeing how many other parts go into Wrapped — not just engineering. There’s a huge team, and we all kind of go together. This was the first time I was working with Data Scientists, Product, and all these other parts that I don’t usually do in my home team. It’s also a global project, so I was working with engineers in Stockholm.
My current role on my home team is Back-End Engineer, and I’m working on the Spotify for Artists part of Spotify. So this is a site and analytics tool where artists go to help manage their career. Right now, my team is focused on creating new APIs for other teams at Spotify. A lot of my days are like the usual Kanban, Agile kind of approach. We try to discuss and figure out things before we build it.”
Here’s what you need to get started
Mindy’s experience working on Wrapped is a great example of how you can always keep learning and growing your skillset as a developer. “Even though I wasn’t a Data Engineer and didn’t have that title, I took the risk to see if I could grow into another opportunity,” she says. “I’m always trying to learn.”
If your company or organization doesn’t have a formal system like Spotify’s embed program, see if there are ways you can get involved in cross-functional projects. It could be a matter of taking someone on another team to coffee and asking questions about their career trajectory or what they’re working on right now. Or you could explore Codecademy projects to get hands-on practice applying your new skills or programming languages.
Feel inspired to learn some of the data skills that are used to code a viral sensation like Spotify Wrapped? A great place to start if you’re new to coding is with the Codecademy path Analyze Data with SQL. You’ll learn how to take large amounts of data and translate it into tables, plus get practice preparing for technical interviews. Another beginner-friendly option that will teach you to use SQL and Python to clean, analyze, and visualize data is the skill path Foundations of Data Science. Be sure to check out the full Codecademy catalog of data science courses and career paths to find something that sparks your interest.