Jasna Jankovic, an associate professor at the University of Connecticut, has spent the last several years developing artificial intelligence tools that help researchers analyze complex microscopy images of materials used in fuel cells, batteries, and other energy systems. Her work represents a quieter but potentially transformative shift in how materials scientists approach discovery, moving away from manual data analysis toward AI-powered automation that can spot patterns humans might miss.

Why Is AI Suddenly Useful for Understanding Materials?

For decades, materials scientists have relied on advanced microscopy techniques to examine the structure of materials at tiny scales. But the sheer volume of data these machines generate has become overwhelming. A single microscopy experiment can produce thousands of images, each containing subtle details about how a material's structure relates to its performance. Manually analyzing all that data is time-consuming and prone to human error.

Jankovic's approach flips this problem on its head. By training AI systems to recognize patterns in microscopy data, her laboratory can now extract insights from images in a fraction of the time it would take a human researcher. "We started working with AI several years ago, before it became widely used," Jankovic explained. "We developed some really unique approaches that we are now publishing and licensing."

These innovations are helping researchers better understand the complex relationships between material structure and performance, accelerating the development of next-generation energy technologies. The practical impact is significant: what might have taken weeks of manual analysis can now be completed in days or even hours.

What Makes Jankovic's Path to This Discovery Unusual?

Jankovic's journey to becoming a leader in AI-driven materials science is anything but conventional. She began her career as a petroleum engineer in Serbia, spending seven years at a refinery before immigrating to Canada in 2002. While raising three children and adjusting to a new country, she pursued a master's degree in chemical engineering at the University of British Columbia.

A turning point came when she learned of an opportunity at the National Research Council of Canada's Institute for Fuel Cell Innovation in Vancouver. The institute was launching a clean-energy research initiative focused on fuel cells, a field Jankovic knew little about at the time. She applied immediately and was hired by a team led by Dr. Radenka Maric, who is now the president of the University of Connecticut. That position introduced Jankovic to materials science and clean-energy research, fields that would define the rest of her career.

"I fell in love with clean energy and with science," Jankovic said. "That ultimately changed my career path." She went on to earn a Ph.D. at UBC while pregnant with her third child, conducting research on proton-conductive materials for intermediate-temperature fuel cells. After completing her doctorate, she joined the Automotive Fuel Cell Cooperation (AFCC), a joint venture between Ford Motor Company and Daimler, in 2011. There, she rose from postdoctoral researcher to senior research scientist, developing advanced microscopy methods that continue to serve as the foundation of her current research.

How Is Jankovic Translating Her Industrial Experience Into Academic Innovation?

When AFCC closed in 2017, Jankovic transitioned to academia at the University of Connecticut. The shift from industry to academia was significant. In industry, she focused primarily on science and a few projects. In academia, she suddenly had to establish a lab, hire students, write proposals, teach, secure funding, manage projects, and build collaborations. Yet her industrial experience continues to shape her approach to both research and teaching.

Her laboratory now develops advanced microscopy and imaging approaches to characterize materials used in fuel cells, batteries, electrolyzers, sensors, and other energy systems. In recent years, her group has become an early adopter of artificial intelligence and automation technologies to improve data analysis and materials characterization. This work has earned her significant recognition in the field.

Steps to Integrate AI Into Materials Research Programs

• Start with data-rich problems: Identify research areas where your lab generates large volumes of microscopy images or other imaging data that currently require manual analysis and interpretation.
• Build interdisciplinary teams: Partner with computer scientists, software engineers, or AI specialists who can help translate your materials science questions into machine learning problems.
• Invest in training and infrastructure: Secure funding for both the computational hardware needed to train AI models and the time for your team to learn new tools and methodologies.
• Start small and iterate: Begin with a focused pilot project on a single type of microscopy data or material characterization task before scaling to broader applications.
• Document and share your methods: Publish your approaches and consider licensing your tools to other research groups, as Jankovic's team is doing, to amplify the impact of your work.

What Recognition Has Jankovic Received for Her Work?

Jankovic's contributions to materials science and AI-driven research have earned her multiple prestigious awards. In 2021, she received a National Science Foundation CAREER Award, which supported her work in advanced microscopy and AI-driven materials characterization. The grant also funded the development of a novel operando microscopy system (a technique that allows researchers to observe materials while they are actively performing their function) and supported numerous student researchers.

"The CAREER Award opened new doors for my research," Jankovic noted. "It enabled many opportunities for both my students and my lab." In 2024, she was honored with the Fraunhofer-Bessel Research Award from Germany's Alexander von Humboldt Foundation, which allowed her to spend six months conducting research at the Fraunhofer Institute for Solar Energy Systems. That collaboration led to the development of a novel approach to X-ray photoelectron spectroscopy for fuel-cell materials.

More recently, Jankovic received the 2025 Women of Innovation Award in Academic Innovation and Leadership from the Connecticut Technology Council. The award recognized her interdisciplinary educational initiatives, including integrating entrepreneurship into engineering curricula and developing STEAM-focused educational projects.

Jankovic's work demonstrates that the most impactful applications of AI in materials science may not be the flashiest ones. Rather than replacing human researchers, her tools augment their capabilities, allowing them to ask deeper questions and accelerate the pace of discovery. As energy technologies become increasingly critical to addressing climate change, these kinds of incremental but meaningful advances in how we understand and design materials could prove essential.