What Does a Data Scientist Actually Do? UVA Alumni Share Their Career Experiences

Data science careers can lead to industries and roles you may not expect. Four University of Virginia School alumni now working in data science explain what they do, how their careers took unexpected turns, and what they’ve learned along the way.

What does a data scientist actually do all day? Is the job mostly coding? And where can a data science degree take you?

A data scientist uses data to understand problems and help organizations make informed decisions. The work can involve building predictive models or analyzing patterns, but it also requires working with people who need to understand and apply the results. How much time a data scientist spends on coding depends largely on the job.

Four M.S. in Data Science (MSDS) graduates at UVA offered a firsthand look at those differences during a recent Modeling Your Future alumni panel moderated by Reggie Leonard, associate director for career connections and community engagement.

The alumni represented very different industries and professional experiences. Some work on established teams, while others are the only data scientist at their organization. Several ended up in careers they hadn't imagined when they entered graduate school.

Their experiences show that data science involves more than technical expertise. It also requires understanding the people and problems behind the data.  

What does a data scientist do every day?

A data scientist's daily responsibilities depend on the organization, but much of the work involves understanding a problem, determining how data can help address it, and communicating the results. Coding may be central to some positions and a smaller part of others.

For Tatev Gomtsyan (MSDS ’24), a data scientist at Citadel Credit Union in Pennsylvania, the job looks different from what she expected.

“I do a lot of machine learning, but it's automated. So, I'm actually not coding day to day,” Gomtsyan said. “I thought I would be doing a lot more technical work, but it is a lot of time spent finding data, using SQL, working with the analysts, getting the data prepared, ensuring its quality, and answering the right questions.”

Gomtsyan works with colleagues across the credit union to understand business needs and develop predictive models. Much of her work focuses on helping the organization better understand its members and how they use financial products.

At TELUS Digital, Jeremy Stern (MSDS ’15) leads an analytics practice supporting digital products for major brands. His responsibilities have evolved from hands-on analytics to managing a team and working directly with clients. For the analysts on his team, the balance between technical and collaborative work is substantial.

“I feel like it is 50% meetings and conversations, and then 50% actual analysis and time in tools,” Stern said.

Those conversations often center on new features being developed for apps. Stern's team helps determine what to measure so the client can understand whether a feature works and how to improve it. The work begins with a question rather than a particular technology.

What industries hire data science graduates?

Data science graduates can find opportunities across industries because organizations of all kinds need people who can turn information into useful insights. The underlying technical skills may be similar, even when the problems differ.

Stern's career took him from biology and digital marketing into analytics leadership, while Gomtsyan originally expected to apply data science to public health. Instead, she found an opportunity in financial services.

“I thought I would honestly take data science into public health. That was my concentration in undergrad,” Gomtsyan said. “Well, I'm in financial services now, so not exactly where I thought I'd end up, and I had zero background knowledge of it. But it worked out. I like it, and I think I had to keep an open mind and commit to learning on the job.”

For Christine George (MSDS ’25) and Kurt Samuels (MSDS ’25), their career paths also took unexpected turns. George studied statistics and political science as an undergrad and worked as a contract data analyst supporting the National Oceanic and Atmospheric Administration (NOAA). Today, she is a data scientist supporting operations at Vivmark Residential, a real estate company that formed from a recent merger between Equity Residential and AvalonBay Communities.  

Samuels has a background in cognitive science and initially pursued technical consulting. He instead joined Pollstar, which serves the live music and entertainment industry.

Although their industries differ, each graduate applies data science to questions that matter to their organization.

Image

What skills do you need to be a successful data scientist?

Successful data scientists need a strong technical foundation, but they also need to understand the context in which their work will be used. An analysis is only valuable if it addresses a meaningful question, and its findings can be understood by the people making decisions.

For Gomtsyan, one of the biggest surprises in her first data science position was how much time she spent explaining the value of predictive models.

“I didn't anticipate the amount of time I'd be spending convincing people that data science works, that these models are helpful, and they'll help us make better decisions and improve outcomes and processes,” she said.

She said that communication is key. “It's kind of like learning to speak a language in a way, how to present these results and translate them in a way that makes sense and that will get adopted by the company.”

That work has required her to think carefully about how predictive analytics fit into the credit union's existing operations. She has spent considerable time documenting processes and helping colleagues understand how models can inform their work.

Stern has experienced a similar need to connect technical expertise with the people who use it. When he joined WillowTree, now TELUS Digital, he expected to work more directly in data science. Instead, he helped establish a new analytics practice that connected technical teams with clients.

“We can speak digital products and we can speak data. We speak marketing and we can speak data,” Stern said. “The data science degree has been helpful to become that glue and that communicator.”

Their experiences reflect an important part of data science education at UVA: Students develop technical expertise while learning to consider how their work will be used and what it means for the people affected by it.

Do data scientists work independently or on teams?

Data scientists may work in large analytics departments or be the only person with that title at an organization. Even when the role is independent, the work usually involves close collaboration with people who bring different expertise.

Gomtsyan knew she would be the only data scientist at Citadel Credit Union when she accepted the position. Even so, the responsibility felt daunting for someone beginning her career.

“I was so scared because, again, limited experience,” she said. “I didn't expect my first role to be something where I'm kind of the only one making the calls and didn't really have anybody else to bounce things off of.”

Although she is the only data scientist, she has since gained confidence and now works with analysts and colleagues throughout the organization to identify problems and determine how predictive models might help solve them.

George works on a similarly small data science team at Vivmark Residential, where analytics supports revenue management and marketing.

“What really drew me to this role was that, because you're in industry, you have more opportunity to collaborate with actual business clients,” George said. “A lot of my work supports our revenue management and marketing teams, so there's been really good opportunity to learn from them and learn the business more, and also learn from really, really smart people.”

Team size can shape a data scientist's experience, especially the independence and opportunities to learn from colleagues. For students considering their first jobs, understanding that the environment can matter as much as the job description.

How did you get your first job in data science?

There is no single path to a first data science job. Applied projects and professional connections can help graduates demonstrate their abilities, particularly when they have limited full-time experience.

Gomtsyan began her job search while completing UVA's MSDS program. She also worked on a research project involving clinical data with the University's Center for Global Health Equity. After graduation, she traveled to Tanzania to complete the project before focusing more fully on her job search.

“It took me, I want to say, five or six months to start my current job,” Gomtsyan said. “The job search was pretty lengthy, but I think that all of my experiences, like MSDS, the capstone, and a project that I did, really helped with building my experience and talking about it when I was networking and doing the job search.”

Samuels found his position through TeamWork Online, an industry-specific job platform that Leonard had recommended.

“You recommended the site TeamWork for me because I was really into sports,” Samuels told Leonard. “I found it on there, just applied, and I got the interviews.”

Image

George's experience illustrates why the first opportunity may not be the right long-term fit. After working as a contractor supporting NOAA, she continued interviewing as she considered what she wanted from her career.

Her search eventually led her into commercial real estate, where she found greater opportunities to work directly with business teams.

“I was not expecting to jump into commercial real estate from the federal government,” George said. “But I stand by the decision, and I think it was a really good move in my career.”

For students, these experiences underscore the value of building a portfolio of work and seeking guidance from career advisers who can help identify opportunities that might otherwise be overlooked.

Can you change industries with a data science degree?

Yes. The skills developed through a data science education can be applied in different industries, allowing graduates to pursue new opportunities as their interests and careers evolve. Moving into an unfamiliar field requires learning its particular challenges, but the technical foundation remains useful.

Samuels originally pursued technical consulting because he wanted a role that combined technical work with opportunities to communicate with clients. “I wanted a little bit more technical skill. That's why I ended up joining the program,” he said.  

Samuels continued exploring consulting opportunities during graduate school before finding his position at Pollstar. “Pollstar is a live music and entertainment publication, which is definitely a very different experience from a government consulting-type role,” Samuels said.

Stern faced a different decision early in his career. He could join an established data science team or help build a new analytics practice at WillowTree.

“It was kind of a decision of being in a place where I'd be one of dozens of people doing the same job and get to learn a lot more from them, but be a small piece of a larger team, and then this opportunity to actually build something from the ground up,” Stern said. He chose the opportunity to build something new.

“I took that jump to try to actually come in and start building something, and I didn't know at the time I would end up still there 11 years later. So, I'm happy with that choice.”

Looking back, Stern acknowledged that the outcome was not entirely predictable. “A little bit of intentional choice and a little bit of luck,” he said.

For prospective students, the alumni's experiences illustrate the flexibility of data science education. Their first jobs did not determine the rest of their careers, and the opportunities they found were not always the ones they originally sought.

Preparing for a Career in Data Science at UVA

At the University of Virginia School of Data Science, students learn to work with data while considering the broader context and consequences of its use. Technical coursework is complemented by applied learning opportunities that allow students to put their skills into practice.

The School's Career Services team helps students explore professional interests and prepare for the transition into the workforce. Through programs such as Modeling Your Future, students can hear directly from graduates about their work and the decisions that shape their careers.

The four alumni offered different perspectives, but their experiences shared a common thread. Data science careers are rarely as predictable as they might appear from a job description.

The work can change as technologies advance and organizations evolve. New opportunities may emerge in industries a student had never considered. What remains valuable is the ability to understand a problem, use data to investigate it, and help others make informed decisions. Graduates can carry that foundation into whatever comes next.

Explore data science degree programs at the University of Virginia.