September 2022 – December 2026
PhD in Computer Engineering
Texas A&M University
College Station, Texas
Thesis: Distribution-guided Parameter-efficient Fine-tuning of Foundation Models for Waveform-based Sensing
About
My doctoral work at Texas A&M University includes adapting vision-language foundation models for histopathology, developing parameter-efficient fine-tuning methods, and modeling physiological time series with physics-informed neural networks. I am also contributing to computer vision evaluation pipelines spanning classification, segmentation, and object detection.
Before my PhD, I designed an integrated wearable stress-detection system during my master’s at Sharif University of Technology, developing both its electronics and machine-learning algorithms.
Education
September 2018 – February 2021
MS in Electrical Engineering
Sharif University of Technology
Tehran, Iran
Thesis: Design and Evaluation of an Integrated Biomedical System for Human Stress Detection and Monitoring
September 2014 – September 2018
BS in Electrical Engineering
University of Tabriz
Tabriz, Iran
Project: Development of a Data Logger for Sensors of Android Smartphones with Python
Research experience
September 2022 – December 2026
Graduate Research Assistant
Texas A&M University
- Developed a distribution-aware method for parameter-efficient fine-tuning of Transformer foundation models for waveform-based sensing.
- Built a few-shot whole-slide image classification pipeline to investigate parameter-efficient fine-tuning of vision-language models to predict ethanol drinking status from liver histology.
- Contributed to an evaluation pipeline for image classification, semantic segmentation, and object detection, comparing pre-trained computer vision models and adaptation strategies.
- Improved the classification accuracy of a convolutional neural network by designing a time-frequency feature combination for acoustic applications.
- Reduced ground truth data requirements by a factor of 15 for physiological signals by developing physics-informed neural networks for cuffless blood pressure measurement.
- Disseminated research findings through publications, presentations, and reports to funding agencies.
July 2019 – February 2021
Graduate Student Researcher
Sharif University of Technology
- Designed an ECG + EDA wearable sensor with a 3× longer battery life compared to alternatives and delivered 94% mental stress detection accuracy across 18 participants using machine learning models.
Teaching & mentoring
- Mentored undergraduate and graduate students on end-to-end AI projects.
- Delivered guest lectures in Machine Learning for Electrical Engineering at Texas A&M University.
- Showcased AI projects at Night at the Zach during the Texas Science & Engineering Fair.
Academic service
- Reviewer, IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).
- Reviewer, Expert Systems with Applications.
- Session chair, IEEE International Geoscience and Remote Sensing Symposium (IGARSS).
Tools & skills
- Coding and computing
- Python, C, MATLAB, NumPy, Pandas, Matplotlib
- Machine learning frameworks
- PyTorch, Hugging Face Transformers, Scikit-Learn
- Development tools
- VS Code, Git, Linux, Jupyter Notebook, High-Performance Computing
- Research interests
- Data Science, Machine Learning, Applied AI, Foundation Models, Deep Learning Architectures & Techniques, Computer Vision, Generative AI, Physics-Informed Machine Learning, Multi-Modal Learning, Efficient AI
Recognition
Fully funded doctoral research assistantship support from MIT Lincoln Laboratory, the National Institutes of Health, and the Texas A&M Department of Electrical & Computer Engineering.
- IEEE Geoscience and Remote Sensing Society travel grant
- Texas A&M Electrical & Computer Engineering conference travel grant
Languages
English (proficient), Farsi (native/bilingual), Azerbaijani Turkish (native/bilingual)