Amanda Fernandez, Ph.D.

Associate Professor
Kay & Steve Robbins Fellow
MATRIX AI Thrust Co-Lead
UT San Antonio | Computer Science

SPII 340J

I lead the Vision & AI Lab at the University of Texas at San Antonio, where my team develops efficient, interpretable deep learning for applications across the sciences, from materials characterization to nuclear security and multimodal data fusion. I serve as Thrust co-Lead for Machine Learning & Deployment in the MATRIX AI Consortium, coordinating research across groups and moving AI toward real-world capability. Our work is supported by the National Science Foundation, the U.S. Department of Energy, and the U.S. Department of Defense, in collaboration with Los Alamos and Sandia National Laboratories.

My approach to research is shaped by a decade in industry as a machine learning researcher and enterprise software engineer before I joined academia; I hold 22 U.S. patents and am a Senior Member of the National Academy of Inventors, the IEEE, and the ACM. Alongside my research, I build programs that bring more students into computing research and mentor early-career researchers - work recognized by the Council on Undergraduate Research's Mid-Career Mentor Award. I am an Associate Professor of Computer Science at UT San Antonio and completed my Ph.D. at the University at Albany, State University of New York.

Research Projects

Research Statement:

My research program advances deep learning that is efficient, interpretable, and trustworthy enough for the domains where errors carry real consequences: national security, energy, and the physical sciences. As models grow more capable and simultaneously more opaque, the binding constraint is no longer raw performance but whether scientists, engineers, and decision-makers can understand, verify, and deploy these systems with confidence.

I lead research at exactly this intersection: optimizing how networks are trained and structured to cut cost and data demands, building attribution and explainability methods that make model behavior legible, and hardening systems against adversarial and real-world failure. I run these as collaborative, mission-driven programs, pairing academic rigor with the deployment discipline I learned in industry, and partnering with national laboratories to move methods from publication to practice. My aim as a research leader is to build the teams and programs that turn frontier AI into dependable infrastructure for science and security.

Teaching