I am broadly interested in the co-design of application and systems - accelerating performance via specializations in algorithms, software, and hardware.
My recent work focuses on building software-defined memory systems to improve datacenter resource efficiency.
I am a Ph.D. candidate at Brown University, advised by Malte Schwarzkopf.
Also, a student researcher at SystemResearch@Google.
Recent Publications:
DaMoN '24
Best Paper AwardIn Situ Neighborhood Sampling For Large-Scale GNN Training
EDBT '24
Crayfish: Navigating the Labyrinth of Machine Learning Inference in Stream Processing Systems
ICS '23
FLASH: FPGA-accelerated Smart Switches with GCN Case Study
I also worked on software & hardware co-design for MLSys training/serving, from FPGA-accelerated storage, network-switches, to stream query processing.
I am a hobbyist of theory and mathematics. I created and taught a course on algorithmic problem solving during 2021-2023 at Boston University (as an extension of the ACM ICPC team and SIAM chapter I founded).
I had lots of fun building HPC systems for cluster competitions during my undergrad, and leading the Massachusetts Green Team (BU, NEU, MIT). media coverage: (1, 2, 3, 4, 5)