About Me

👋 I’m Amir, a Computer Engineering PhD candidate at Texas A&M University. I work on making deep learning models more efficient and more useful in real-world settings — from adapting foundation models with fewer parameters to using physics to reduce the need for labeled data. The projects are funded by MIT Lincoln Laboratory and the National Institutes of Health. I have published in Nature NPJ Digital Medicine, IEEE venues, and WACV, among others.

My research spans audio, image, and time-series modalities, with applications in acoustics and health. These are all areas where data is noisy, labels are expensive, and you cannot just train a bigger model. I started in hardware — during my master’s at Sharif University of Technology, I designed a wearable sensor from the circuit board to the classification algorithm in software. That experience taught me to think about the whole problem, not just the model. I carry that into my current work: understanding the signal, the physics, and the constraints before deciding what to build.

🎓 Education

Doctor of Philosophy, Computer Engineering
Texas A&M University
Sept. 2022 – Present (anticipated December 2026)

Master of Science, Electrical Engineering
Sharif University of Technology
Graduated February 2021