Typical day in the life of a data scientist manager

Figueroa: As a manager now, I do less coding than I used to. Previously, a typical day started with morning stand-ups with my team. The first thing we did in the morning was a lot of data wrangling and coding, and we needed to understand what we needed to do — for example, building models.

Half the day, we built these models, brainstormed how our work might impact our client's business, and presented those results. So it's part research, part engineering, and obviously, part communication.

Figueroa: In a manager role, it's different. You're basically the bridge between the business and the technical teams. At Xtillion, I'm leading the stand-up meetings, managing the logistics of what we need to do, and making sure the whole team is focused on our goals because they will yield certain results. I also offer guidance and mentoring in terms of implementation. It's mostly about making sure the analysis we're doing, the data we're facing, is done in an efficient manner, and that we're yielding the correct results at the end of the day.

Figueroa: As you grow in your data science career, you'll have to develop a lot of soft skills. Most of the time, engineers get caught up in numbers, models, and graphs. But, when communicating with leadership, the first question they usually have is, Why does this matter for my business? They're less concerned with the model specifics and more interested in Is it accurate? How will it impact my business? Will it reduce costs?

I also took the time to get mentorship in communication skills — verbal, written, presentation — on the side to prepare for this kind of role. Conveying complex terms in a simple, clear manner is so important.

It's up to me to sit down with the client and explain, I know you want to build this, but if we want to set you up for future success, we need to address these foundational elements. It's about translating all that technical expertise into layman's terms so that the business can understand the necessity of our approach.

Figueroa: It's great. That's one of the things I love about Xtillion. I work a bit too much sometimes, but that's because I genuinely enjoy what I do. Overall, I feel great; I have time for my family, my hobbies, and I even spend time with my team after work. I'm not consumed by my work or burned out.

Around eight hours a day is a normal workday. This also depends on crunch times or when deliverables are approaching. If you're nearing a deadline, you might expect to work nine or 10 hours, but under typical circumstances, it's just an eight-hour workload.

Figueroa: It's very interesting because I know of different profiles within Xtillion. They're all data scientists, but they took different paths.

Figueroa: In my case, I have a bachelor's degree in computer science and a master's degree in computer engineering focused on machine learning and deep learning from the University of Puerto Rico.

So that's a traditional case: a bachelor's degree in computer science or software engineering, and then jumping to a master's degree in data science, statistics, or math.

Figueroa: Another common path I've seen in the firm is that of industrial engineers who jump into data science. They do very well because industrial engineers need to understand organizations and optimize processes, which is basically data science at some point.

It mostly depends on what you're expecting your work to be. If you want to become a data scientist who is embedded in the business, you need a profile that reflects that. If you're a data scientist who focuses on infrastructure and algorithms, then you need foundational knowledge of computer science.

Figueroa: I know data science bachelor's degrees are starting to pop up more, so I'm very interested in seeing how they do, especially given the rigorous math and coursework involved.

Figueroa: I don't think there's a shortcut to becoming a data scientist. As every engineer will tell you, there is a lot of math involved in this career — advanced math. You have to take your time to understand it. You know how we used to say in school, Why am I learning all this? Does it even matter? Well, in data science, it does matter. It's math that you work with every day.

Even though it is not often talked about, communication skills are invaluable for translating all of your math and technical skills for clients and stakeholders. If you can communicate clearly and confidently, that will take you far.