Differential Privacy
Provable guarantees that protect individuals in a dataset. I design DP algorithms that stay accurate at scale - on billion-edge graphs and streaming queries.
I turn mathematically rigorous ideas - across privacy, distributed systems, and machine learning - into practical software with measurable performance gains.
Algorithms · Differential Privacy · Distributed & High-Performance Computing · ML Systems · C++/CUDA
I am a Ph.D. candidate in Computer Science at Yale University and an Applied Researcher Intern at eBay. At Yale I am part of the WildAlgs Lab, advised by Professor Quanquan Liu. My research sits at the intersection of algorithms, differential privacy, distributed and high-performance systems, and machine learning.
I have developed distributed differentially private graph algorithms with substantially improved empirical accuracy, learning-augmented methods for private online analytics, scalable semantic-search systems, neural architectures for mathematical reasoning, and GPU-accelerated scientific software. My work has appeared at venues including VLDB, ICDE, and SIGMOD, with research experience across Yale, MIT CSAIL, Caltech, the University of Rochester, and the University of Washington.
Five areas run through my work.
Provable guarantees that protect individuals in a dataset. I design DP algorithms that stay accurate at scale - on billion-edge graphs and streaming queries.
Splitting work across many machines. I build parallel, communication-aware algorithms that scale to massive graphs.
The infrastructure that trains models at scale. I engineer distributed graph-learning and GNN pipelines for recommendation.
Finding structure in large, messy data. I work on semantic set search, coreset selection, and approximate query answering.
How data is stored, queried, and discovered. I work on query models, schema and dependency discovery, and high-performance access.
Recent peer-reviewed work.