// cv
MComp student at the NUS School of Computing, with a background in GPU systems engineering and five years of industrial embedded development.
Currently completing my capstone project under Prof. Bingsheng He, alongside research internships in Prof. Shuicheng Yan's LV-Robotics Lab and the G&M Lab.
My research interests center on multimodal (vision-touch) world-action models for robotics, and on-device inference optimization.
Download PDF Resume// skills
Programming
Tools
Hardware
Languages
// research interests
World Action Models & Vision-Language-Action
Humanoid Robot Systems
Efficient Inference for Real-Time Policy Execution
// timeline
Aug 2025 — Present
Master of Computing (General Track)
National University of Singapore, School of Computing
- Expected graduation 2027
- Coursework: CS5242 Deep Learning, IT5005 Artificial Intelligence, IT5003 Data Structures and Algorithms, CS5224 Cloud Computing
Jul 2022 — Dec 2024
Intermediate Embedded Hardware Engineer
Lead Intelligent Equipment Co., Institute of Technology
- Independently designed ARM-MCU-based industrial products: RS485 communication terminals, a proportional-valve PID controller, and detection/control terminals for lithium-battery formation and aging
- Served as project manager, coordinating cross-functional resources; products passed CE/TUV certification
Jul 2020 — Jun 2022
Project Manager & Hardware Engineer (promoted from Intern Hardware Engineer)
SUPCON Technology Co., PLC Product Sub-Center
- Led the end-to-end design and productization of the minimum system control board for PLCs based on domestic Loongson CPUs
- Developed analog voltage acquisition, RTD measurement, and digital I/O modules for PLC and DCS product lines
Sep 2016 — Jun 2020
Bachelor of Science in Automation (Electrical)
Zhejiang University, College of Electrical Engineering
// project experience
Robot World Video Decomposition Model
Evaluating single-image layer decomposition methods to build the foundation for a tri-view layered World Action Model in robot manipulation.
MaxQ: a universal back-end compiler parameter optimization system
A universal back-end compiler parameter optimization system.
G-Watch: a GPU Profiler with AI-Driven Automatic Optimization
A CUPTI-based GPU profiler for LLM inference on SGLang and vLLM, exposing a Python API for kernel-level bottleneck analysis.
Marine Life Classification with Deep Learning
An end-to-end image classification pipeline on 20k+ underwater species images, comparing ViT and EfficientNet-B0.