Andy Huang turned a broad interest in AI into an award winning thesis and a first job in data science

A thesis on dementia does not sound like the obvious route to a job at Albert Heijn. For JADS alumnus Andy Huang, it became both: a Tilburg University award for Best Master’s Thesis and a strong start to his career as a data scientist. That is part of what makes his story interesting for bachelor students thinking about their next step. Andy did not begin with a perfectly mapped-out plan. He finished his AI bachelor at Radboud University Nijmegen feeling interested but not fully grounded yet. JADS gave him the place to turn that broad interest into something more practical, focused, and useful. Discover Andy’s story and how he won the Master’s Thesis Awards on March 2026.

How it all started

“I was unsure whether to pursue a master’s or start working right away,” Andy says. “Part of me felt that gaining hands-on experience first might be the better next step.” Instead, he chose the Master of Data Science in Business and Entrepreneurship at JADS. Looking back, he is still happy with that decision. Winning the thesis award made that choice visible in a new way. For Andy, it was a personal surprise. For Dr. Georgios Manias, one of his supervisors, it was also a sign of how strong student work can become when technical depth, structure, and relevance come together.

“What stood out was the clinical impact,” Georgios says. “Andy managed to build on earlier studies in a strong way and contribute something meaningful for follow-up research and practice.”

Andy Huang, JADS alumnus, received the Master’s Thesis Award from prof. dr. Wim van de Donk, Rector Magnificus and Executive Board of Tilburg University

From broad interest to real application

Before JADS, Andy studied AI. He liked the mix of subjects, from neuroscience and psychology to computer science and mathematics, but he also felt something was missing. “The bachelor was broad,” he says. “I liked that, but afterward it was still hard to connect everything. It was not tangible enough for me.” That is where JADS stood out. For Andy, the program offered more than technical content. It showed how data science connects to real questions, real organizations, and real people. “JADS gives you a solid theoretical foundation,” he says. “What makes it different is the strong focus on application: learning how to translate technical knowledge into practical solutions that create value in the real world.”

He also remembers the atmosphere. JADS felt small in a good way, with a close community and a lot of students who were ambitious, creative, and willing to build something of their own. “It is a vibrant place,” Andy says. “Many students are creative and ambitious. People are building things, exploring startup ideas, or already launching them. Being surrounded by that energy really motivates you to aim higher.”

A thesis that had to work in practice

Andy’s thesis focused on the early detection and prevention of dementia. It is a complex topic, but his goal was clear from the start: build something that could help in practice. “Dementia is a big societal challenge,” he says. “Diagnosing dementia is difficult. It does not rely on a single test, and there are many intercorrelated risk factors.” That matters because dementia often develops before symptoms are visible. Andy’s research explored whether data science could help identify risk at an earlier stage, when there is still room to act. “We wanted to explore whether it is possible to predict the onset of dementia before actual symptoms appear,” he says. “If you can do that, you can create more time and opportunity to intervene.”

What made the thesis stand out was not only accuracy, but explainability. In healthcare, a model needs to be understandable as well as strong. “A model that is not interpretable is not useful in the medical domain,” Andy says. “Clinicians need to understand what drives a prediction.” His approach was also broader than many earlier studies. Instead of focusing on one type of data, the research combined clinical evaluations, neurological assessments, lifestyle factors, and other inputs. That made the model more holistic and closer to the real complexity of dementia. For Georgios, that broader setup mattered as much as the technical work itself. “This was not a project built in isolation,” he says. “It connected technical research to a real healthcare context, and that made the results much more valuable.”

Dr. George Manias, Assistant Professor, Data Governance and supervisor of Andy

Learning by stepping into a new field

One of the strongest parts of Andy’s story is that he did not begin with a medical background. He had to build that knowledge during the project. “I had never worked in the medical domain before,” he says. “So I was missing a lot of domain knowledge.” He filled that gap by reading extensively and by working closely with clinicians. That combination helped him understand both the research and the meaning of the data behind it.

The project was linked to COMFORTage, a European research project on dementia and frailty, and used data from Eginitio Hospital (from the National and Kapodistrian University of Athens) involved in that work. That made collaboration an important part of the process from day one. “We were using data from the hospital, and the results were meant for clinicians,” Andy says. “So to me, it made sense to involve them directly, rather than just producing something that might not align with their needs or translate into clinically meaningful outcomes.”
Georgios also sees that collaboration as one of the reasons the thesis had wider reach. The work did not only use data from an international setting, it also stayed connected to the questions and needs of healthcare professionals throughout the process. “We wanted the research to reflect the voices and needs of healthcare professionals” he says. “That is why constant consultation with the hospital’s healthcare professionals was conducted. And that we worked in a broader European and multidisciplinary context.”