Engineering, data science and artificial intelligence: Juan Segundo Hevia's experience at Rice University

Juan Segundo Hevia is 30 and graduated in 2025 from the master's in Data Science at Rice University, in Houston. Before starting graduate study he had already worked several years in data science and machine learning, both at companies and on projects tied to artificial intelligence and public policy, and he had taken part in academic research within his field.

From industrial engineering to data science

When he finished secondary school, choosing industrial engineering was tied at first to the subjects that interested him most and the ones he felt most comfortable with.

"I knew I really liked math, I did very well at it, and I liked playing with numbers and with physics too. That led me to industrial engineering."

He studied on a scholarship and, as he advanced, he started taking a particular interest in statistics, programming and data analysis, an area that at the time still had less presence inside traditional engineering curricula.

"I saw that the whole data and statistics side had a lot of potential ahead of it. I did my first internship and that's how I got to my career in data science."

The industrial engineering training gave him a quantitative base in math, statistics and optimization that he could later carry directly over to data science. Starting with his first jobs, that orientation began to take up more and more space in his professional career.

Professional experience before the master's

His first internship was at Clarín, where he worked as a data scientist. Later he began coordinating programs at Universidad Austral that connected the fundamentals of artificial intelligence with governance and public policy, aimed mainly at professionals interested in understanding how to apply these technologies in the public sector.

In 2020 he joined Memorable AI, where he held the roles of machine learning engineer and, later, head of data management. The company developed technology for digital advertising using models capable of analyzing and predicting different characteristics of creative assets, and Juan worked there on computer vision systems and on the infrastructure needed to manage a growing volume of information. He stayed at the company until 2023, when he left to study at Rice University with the Fulbright scholarship, and in 2024 Memorable AI was acquired by Reddit.

By the time he started the master's, then, his interest in data science was already backed by several years of technical work in industry, rather than by a decision to specialize taken only on arriving in the United States.

The Fulbright process

Juan submitted his Fulbright application on March 12, 2022 and started classes at Rice in August 2023. About seventeen months passed between those two dates, because the scholarship process starts well ahead of the beginning of the academic year in the United States.

"March 12, 2022. I still remember finishing the whole form."

When he began the application he was fairly clear about the kind of program he wanted to do, though he had not yet settled on a single university.

"I knew I wanted to do a master's in Data Science. I wasn't very clear on which university, though obviously I had my preferences."

The Fulbright process also led him to widen the list of institutions he was considering and to research programs that were not originally among his first options. Rice was one of them.

"It helped me a lot to learn about a whole lot of other universities, like Rice, which wasn't my first option, but which I learned is one of the leading engineering universities in the United States."

For anyone considering a similar call, the timeline matters because preparation usually starts more than a year before the trip. Beyond the scholarship, you then have to complete the university applications and wait for each institution's admission and funding decisions.

How he funded the master's

In Juan's case, the funding combined Fulbright's support with a scholarship granted by Rice. The university ended up covering 100 percent of his tuition, while Fulbright contributed other components tied to the experience of studying in the United States.

"When they told me I got in, what happens is that Rice also covers my tuition, which is a very important part of the cost of living here. I had 100 percent coverage, so it helped me a lot."

It's important to separate the two sources because they follow different processes. Fulbright and the university may contribute different components of the budget, and the final package depends both on the external scholarship and on the financial offer from the institution that admits the student.

In his case, that combination meant Rice's tuition was covered in full. Juan remembers getting the news while he was at home in Buenos Aires and that his first reaction was to call family and friends.

The Data Science master's at Rice

Rice's program runs inside the engineering school and is designed as a professional master's, so it doesn't require a traditional thesis. The structure includes between 31 and 35 credits, with required courses in programming, statistics, visualization, machine learning and data science tooling, plus electives that let students steer the program toward different areas.

Students can go deeper, among other options, into artificial intelligence, business analytics, the energy transition, image processing or sports analytics. The program ends with an applied project in which students work on a concrete problem alongside an outside organization.

For Juan, one of the advantages of that structure was being able to fit the courses to his technical interests and combine more general content with advanced topics in artificial intelligence.

"It lets me choose a lot of the profile I want to take on, from the most basic programming to the most complex generative artificial intelligence. I can build the profile I want, and I enjoy that a lot."

His final project was done with NASA and focused on lunar crater detection for the Artemis II mission, a computer vision application that let him work on a problem different from the ones he had previously run into in advertising or public policy.

From a course to a publication

One of the courses that interested him most during the program was Advanced Machine Learning, where he worked on applications of language models in education. The initial course project later continued outside the classroom and ended up becoming an academic paper.

"That later helped me expand the work from that course into a paper I'm presenting at the Association of the Advancements in Artificial Intelligence conference."

The paper, Towards an Efficient, Customizable, and Accessible AI Tutor, was written together with Facundo Arredondo and Vishesh Kumar and presented in March 2025 at a workshop on artificial intelligence and education in Philadelphia. The work was also published in the Proceedings of Machine Learning Research.

The case shows one of the ways a professional master's can connect with research: a project developed initially for a course can continue after the semester if there's a question worth exploring in more depth.

During those years Juan also took part in another published work together with Santiago Segarra, Lisa O'Bryan and Madeline Navarro in the Journal of Personality and Social Psychology, which used machine learning tools to analyze and predict speaking turn patterns within work teams. In that way, his academic output during the master's ended up covering fairly different problems, from computer vision to education and organizational behavior.

Living in Houston

Life outside the university was also shaped by the community of Argentines he found in Houston, a group that formed around everyday activities like playing soccer and that ended up working as a network for sharing the adjustment to living in another country.

"My life in Houston involves a lot of heat. It's a very hot city, but the people are also very warm. There are a lot of us Argentines and that makes me feel there's a bit of home in this new home."

For Juan, having that community mattered especially during the first months, when many of the difficulties had nothing to do with coursework but with more practical matters, from organizing daily life to understanding how certain financial and administrative things worked in the United States.

"I think that's the key: having a group, a community to share things with. And if you don't have one from the start, building it."

Professional networks and what came next

When he talks about what he'd recommend to someone thinking about studying abroad, Juan puts a lot of emphasis on meeting people already working in the areas that interest you and on using the university as a space to understand what projects, companies and research are being developed around you.

"The value of people, the value of meeting people, and being proactive about it. Studying here is my first priority, but it's not the only thing I do, and a big part of the rest of what I do is looking for what people here are doing: what the Argentines are doing, what the Latinos are doing, what people in the United States are doing."

Rather than thinking of networking as an activity separate from the training, he describes it as a way of exposing yourself to different jobs and paths, asking how others reached certain roles and understanding what possibilities exist within a field that changes fast.

"I think it's a fairly flexible rule and very applicable to any setting: go out and look for the people who are working, ask, be curious, and try to connect and learn from the other person."

After graduating in 2025, Juan founded HeyBeta out of a simple idea: artificial intelligence can already change how we work, but most people still don't know how to use it. HeyBeta exists to close that distance, building tools that let anyone, with no need to know how to code, bring AI into their daily work. For Juan, that's where the real impact of the technology is: not replacing people, but giving them back time and multiplying what they can achieve.