cv

Ph.D. Candidate

Washington University in St. Louis, St. Louis

Computer Engineering

• Advisor: Professor Chenyang Lu

Bachelor of Science in Engineering

University of Michigan, College of Engineering, Ann Arbor

Computer Engineering

Exchange

Instituto Tecnológico de Buenos Aires, Buenos Aires, Argentina

Exchange

Technische Universität Berlin - TU Berlin, Germany

Bachelor of Science in Engineering

Shanghai Jiao Tong University, UM – SJTU Joint Institute, Shanghai, China

Electrical & Computer Engineering

Best Student Paper Award

IEEE Real-Time Systems Symposium (RTSS'23)

[Paper] Progressive Neural Compression for Adaptive Image Offloading under Timing Constraints

Summer Research Intern

Washington University in St. Louis

Department of Electrical and System Engineering with Prof. Bruno Sinopoli and Dr. Sankalp Bhan

• Developed new course materials about linear control, system identification and linear state estimator for Washington University in St. Louis.
• Independently developed model reference control and adaptive control algorithms for a parrot quadcopter.
• Realized Indoor attitude and position tracking and estimation with OptiTrack infrared camera system.

Optimizing Edge Offloading Decisions for Object Detection

2024 IEEE/ACM Symposium on Edge Computing (SEC)

The paper introduces a reward-based metric for offloading decisions to maximize object detection accuracy under resource constraints on embedded devices.

Real-Time Human Action Recognition on Embedded Platforms

arXiv preprint arXiv:2409.05662

This work addresses real-time human action recognition (HAR) challenges on embedded platforms by designing a novel motion feature extractor and real-time system achieving 30 FPS on Nvidia Jetson Xavier NX.

Progressive Neural Compression for Adaptive Image Offloading Under Timing Constraints

2023 IEEE Real-Time Systems Symposium (RTSS)

Proposes Progressive Neural Compression (PNC) to adaptively offload images over variable wireless bandwidth, optimizing inference performance under timing constraints. Received Best Student Paper Award.

Contact Tracing for Healthcare Workers in an Intensive Care Unit

Proc. ACM Interact. Mob. Wearable Ubiquitous Technol.

Presents CATCH, an automated Bluetooth-based contact tracing system tailored for healthcare workers in ICU environments, demonstrating superior accuracy and efficiency.

Adaptive Edge Offloading for Image Classification Under Rate Limit

IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems

Proposes a lightweight DQN-based offloading policy for image classification that optimizes accuracy under rate-limiting constraints using a token bucket mechanism.