That vagueness leaves some people wondering where the path to becoming a data scientist begins and what they need to know about the burgeoning field. While a direct route to a career in data science may begin with a background in the field, including a master’s degree from one of dozens of data science programs, other people take a more circuitous route to this role.

Advertisement

Syracuse University MS in Applied Data Science Online

Our Updated Curriculum at Syracuse University Prepares You to Be a Leader in Data Science

Don't get left behind as the data science industry evolves. In as few as 18 months, earn your MS in Applied Data Science from Syracuse University, ranked #25 in Best Online Graduate Computer Information Technology Programs. Apply today to gain the analytical, technical and managerial expertise required to stand out in a competitive job market. No GRE required.

Read More

Learn More

Learn More

• GRE RequiredNo
• Part TimeYes

That was the case for Eugenio Zuccarelli, now the manager of data science for CVS Health. After earning a bachelor’s degree in electronic and information technology engineering at the University of Genoa in 2012, he went on to pursue a master’s degree in biomedical engineering, which sparked his interest in machine learning. Zuccarelli ultimately completed a master’s degree in business analytics from Massachusetts Institute of Technology’s Sloan School of Management in 2019, which helped him become a data science leader at one of the largest companies in the world.

Whereas realms of data science like artificial intelligence (AI) technology were largely still considered science fiction when Zuccarelli finished his undergraduate studies, he says that developments in machine learning and AI are now at the forefront of the minds of many industry leaders. In his role at CVS Health, ranked No. 4 on the Fortune 500 list, Zuccarelli leads innovation efforts to develop AI technology in the health care sector.

Fortune sat down with Zuccarelli to learn more about his experience in the burgeoning world of data science, and how up-and-coming data scientists can succeed in this rapidly changing field.

To apply data science in an ‘effective’ way requires a grasp of the domain

It was while completing his first master’s degree—a biomedical engineering program with a distinction in neurotechnology at Imperial College—that Zuccarelli first became interested in how machine learning could be applied to find practical solutions. His thesis project entailed developing a robotic prosthesis: an engineered arm controlled by the thoughts of a user. 

“That really propelled me into the specific area of AI—but specifically into something that’s going to have an impact on people,” he says. “I was working with people that have had severe trauma, and so this relationship or this human component became extremely important for me and kind of led all of my work after.”

After working in roles as a data analyst and a data scientist, Zuccarelli went on to complete his second master’s degree in business analytics in 2019 from MIT. By then, data science and AI—and their applications in the business world—had grown immensely, he recalls.

Zuccarelli learned more about the potential of the intersection between data science and health care through research projects he conducted with MIT between 2020 and 2021. There, he was able to develop predictions using information about U.S. COVID-19 deaths and hospitalizations as a part of a task force working with the White House.

“When you have a grasp of the domain of the industry and how it can be helpful, you can really apply data science in an effective way,” Zuccarelli says.

The jump from junior data scientist to data science leader

While complex technological developments are of interest to many aspiring data scientists, the vast majority of data scientists use very little AI and machine learning technology at all.

“Some of the junior people come from school into a new role and they tend to think: ‘How can we involve something like deep learning? How can we make a very complex and interesting technique for me to learn and put into practice?’” Zuccarelli says. “But most of the time, this is not actually what the stakeholders want, what our clients want, or what the business can support.”

What is focal to leading a data team successfully though, is the ability to effectively and meaningfully communicate insights. This becomes all the more essential as a data scientist climbs the ranks. Additionally, leaders should understand how other, non-data-informed factors may play into their executives’ decision-making.

There are times when leaders may decide to not take action based on the data and pursue a different strategy, Zuccarelli says. But that’s not necessarily an issue by itself because sometimes there are questions about the quality of data and therefore, how strong the conclusions drawn from it actually are.

Aside from seeking simpler solutions, as with a lot of jobs, senior data scientists tend to shift from focusing on technical skills to thinking more holistically and strategically. However, data science leaders should still maintain their technical and coding skills so they can understand what is feasible for their team. 

“For me, that transition went from having to do a task or having to implement the feature to having to envision a product and then seeing if we can apply data science to it in a way that’s going to be effective,” Zuccarelli says. 

Data science leaders must know when to implement more complicated solutions

The ability to reliably predict the outcomes of a heart surgery using modeling may have been a pipe dream a few decades ago, but solutions like these are now at the crux of the intersection of technology and health care. 

When consulting with industry executives, data scientists will comb through mountains of data and try to find insights, then they will surface key takeaways and actions to executives, Zuccarelli explains. AI and machine learning are useful tools in cases where there’s a need for predictive analytics, he continues. 

“There have been a lot of research shortages on predictive analytics—how can we predict diseases, complications, surgery outcomes—and that’s really sad, certainly because it’s so close to the human experience,” Zuccarelli says.

At MIT, Zuccarelli’s primary research involved predicting whether a child will survive surgery, specifically congenital heart surgery. In these cases, there’s a child who has clear malformations and was born with a heart that was not functioning properly.

“The doctors need to do surgery but they might not know, of course, what is the best surgery to do on this person, and whether they will survive in the future,” Zuccarelli says. “And that’s where machine learning comes into play—they can make predictions based on a lot of past data.”

Such hands-on research is not only helpful for developing technical skills and processes, but has broader applications.

“It is one of the greatest ways to interact not just with data and with technology, but also with people,” he says. “We’ve worked a lot with doctors and also gained the ability to communicate well with non-technical stakeholders, doctors, and business people about data science.”

Data science and AI can be used as a tool—not necessarily a replacement—for doctors and healthcare industry leaders like it is in other business landscapes. 

“My argument is that machine learning is a bit too much when it’s for something a doctor could do, but in a lot of situations, machine learning and AI can make predictions,” Zuccarelli says. “People tend to struggle, of course, to make predictions for the future.”

Check out all of Fortune’s rankings of degree programs, and learn more about specific career paths.