Home

I am an Assistant Research Professor in the Department of Biostatistics & Bioinformatics at Duke University. Before joining Duke, I spent over four years as a Research Scientist in industry, followed by a postdoctoral appointment at the University of Pennsylvania. I received my Ph.D. in Electrical Engineering from the University of Southern California in 2016, my M.Sc. from Cornell University in 2014, and my B.Sc. in Electrical Engineering from Sharif University of Technology in 2011. I also spent the summer of 2015 as a research intern at Bell Labs.

My research bridges artificial intelligence (AI), machine learning (ML), and computational biology to develop mathematically grounded frameworks for learning, prediction, and design in complex biological and clinical systems. I combine tools from optimization, signal processing, and optimal transport to create efficient, interpretable, and scalable AI models that move beyond passive prediction toward AI-driven molecular design and clinical decision-making. My group’s recent work includes representation learning for protein language models (PLMs), including in the context of de novo peptide sequencing in mass spectrometry, reinforcement learning (RL) for molecular and clinical design, and constrained multi-objective learning for protein engineering. Through collaborations in radiology, proteomics, and healthcare, I validate these methods on real-world datasets, advancing applications in proteomics, drug discovery, and precision medicine.

Earlier in my career, I worked extensively on network information theory and wireless communication systems, studying how distributed systems exchange and optimize information. These experiences continue to inform my current work, connecting theoretical foundations with biological complexity to develop next-generation AI systems for life and health sciences.

What’s New

• Jan. 2026: I am serving as an Early Career Reviewer for the Biodata Management and Analysis (BDMA) Study Section at the National Institutes of Health (NIH).

• Dec. 2025: Two new preprints:

  • LUNA: Linear Universal Neural Attention with Generalization Guarantees

  • Efficient Transferable Optimal Transport via Min-Sliced Transport Plans

• Oct. 2025: Three new preprints:

  • Protein Language Model-Aligned Spectra Embeddings for De Novo Peptide Sequencing

  • Reinforcement Learning-Based Optimization of CT Acquisition and Reconstruction Parameters Through Virtual Imaging Trials

  • LOTFormer: Doubly-Stochastic Linear Attention via Low-Rank Optimal Transport

• Aug. 2025: I am serving as an Area Chair for the NeurIPS 2025 Workshop on Constrained Optimization for Machine Learning.