When Siphu Langeni describes her journey from working in healthcare to working in commerce as a data scientist, she tells people: “I went from saving lives to saving livelihoods.”
Siphu is a data scientist at Shopify, the platform that merchants and entrepreneurs use to start, grow, market, and manage in-person and online businesses. Before switching to tech, she worked in a clinical setting as a Certified Registered Nurse Anesthetist.
A major life event, plus “a healthy combination of confidence and a little bit of crazy,” prompted Siphu to start over and enroll in a data science certificate program. She dove into her data science coursework and watched video tutorials about coding in her spare time. “I felt like I couldn’t get enough,” she recalls.
Siphu landed an internship with Shopify in the summer of 2021, which gave her hands-on industry experience, and ultimately led to a full-time job as a data scientist in Shopify’s commerce algorithms and product understanding department.
Starting over in a new industry was tough initially, but it was a much-needed change, Siphu explains. “Everybody is job-seeking — and that’s cool,” she says. “But when you’re looking for a career, you need a thing that’s going to keep you coming back every day. I felt like the problem spaces within data science were so impactful.”
Curious what a career in data science could look like for you? Check out our skill path Data Science Foundations. It’s a beginner-friendly course that will teach you how to clean, analyze, and visualize data responsibly, plus introduce you to some of the jobs you can get in data science. Here’s how Siphu learned how to use code in data science, found a mentorship program, and ultimately got hired as a data scientist at Shopify.
What got me interested in the job
“I wanted to do something different, but I didn’t know what that was necessarily. Working in healthcare, especially in very high acuity areas, there’s a tendency to run into burnout, so I knew it was a very real risk. I was at a point in my career where I felt really good; it’s comfortable to be in a place where people see you as kind of an expert at what you do.
I was getting a lot of exposure to decision-makers and policy-makers, and I found that I always seemed to provide the most compelling story or argument when I was able to back it up with data. I was incorporating a lot of data, but I wasn’t calling it ‘data’ at that point.
I decided to go back to school full time and study for an advanced diploma in data science. The stars aligned where the opportunity presented itself, and I was mentally ready to be in a position of starting over — because it’s tough.”
How I got in the door
“Imagine moving from one completely different space to another. There’s a lot of ambiguity, a lot of questions, a lot of blanks that need to be filled in.
After leaving school, I needed something that would allow me to be more industry-ready and employable. One of the ways I was able to do that was through a mentorship program called SharpestMinds. I was crawling going into that, and when I came out of it, I was running.
I was paired with a mentor working in the industry, and we were able to produce an end-to-end data science project. The project was related to the NLP (Natural Language Processing) space, so using language to predict certain things. I needed annotated data, so in the process, I actually created and authored my first Python package. That project let me learn a lot of technical skills and build confidence.
I started at Shopify as an intern, which ended up being a really great experience for me. It gave me hands-on experience to better understand the role of data science and what it even is. Fortunately, my team thought I was pretty good, so they kept me and I stayed on board. I’m actually on the same team that I interned on.”
What I actually do every day
“If you ask 10 data scientists what they do, they’ll all give you 10 different answers. One of the reasons why Shopify was one of my top choices is because I did want to be practicing data science and be very close to the coding side of things.
I love trying to figure out a problem, then translating that to code, pushing it to production, and actually seeing it do something cool. I feel really great when I’m working on coding.
The team that I’m on specifically is called ‘leverage,’ which focuses on giving our merchants the ability to leverage the scale that Shopify has and that network effect that they could never otherwise have on their own. That involves working very closely with product, engineering, and UX, building products that help make it easier for merchants to do that.
One example of something we’ve worked on is a product classification model. This is a really cool machine-learning model that inputs text and image features, and is able to categorize products appropriately. In order for businesses to have good SEO, to be discoverable, and to have proper taxation, their products have to be categorized correctly. This is one of those things that we provide to make it easier for merchants to be able to get their business going.”
Here’s what you need to get started
Like lots of data scientists, Siphu primarily uses Python as her main programming language. Working with software engineers, Siphu realized that she writes “production-grade code” that’s not reusable without solid documentation. “Good documentation has saved me on many occasions,” she says.
Shopify is all about empowering entrepreneurs and using data to level the playing field. With billions of products sold on its platform, Shopify works with a ton of data — “petabytes worth,” according to Siphu. So she uses SQL to access and manipulate the large volume of data. A strong understanding of SQL is very valuable as a data scientist, because you’re often using advanced queries to access data. “Data isn’t useful just by nature of it being there,” she says. “You have to be able to turn it into something that you can pull insights out of.”
Want to learn how to use Python and SQL like a data analyst? In our Career Path Data Scientist: Analytics Specialist, you’ll learn how to “talk” to data, make visualizations to see the big picture, and conduct tests. You’ll also walk away with some portfolio-ready projects under your belt that you can use in job interviews.
Beyond those technical skills, data scientists also have to be able to communicate and contextualize their findings. “So you find all the secrets that are locked away in the data — what if no one has any actionable insight or anything they can do as a result of what you’ve discovered?” Siphu says. “Then you’ve just had fun with the data, which is really cool, but then what?”