Hi, I'mHamster.
A
Self-driven, quick starter, passionate researcher with a curious mind who enjoys solving complex and challenging real-world problems.
About
I am a Second Year Master Student at University of Michigan - Shanghai Jiao Tong University Joint Institute. I enjoy problem-solving for the fields of time-critical computing and coding, of course! Always strive to bring 100% to the work I do. In my undergraduage study, I have worked on technologies like C/C++, Python, Shell, Cuda, MatLab, MySQL, Java during research and course projects. I have 6 months of professional work experience which helped me strengthen my experience in Python and Shell. I am passionate about developing complex applications that solve real-world problems impacting millions of users.
Research Interests:
Skills Summary:
Looking for an opportunity to work with passionated collaberators with relative fields, such as computer architecture, end-to end scheduling, ML system, efficient AI, low power design, autonomous system, and so on!
Publication
Y. Shen, Y. Ni, J. Zhu, J. Zhao, and A. Zou, “Compression Space Search: RL-Based Combinational Compression for Neural Networks,” 2026 Design, Automation & Test in Europe Conference & Exhibition (DATE), Verona, Italy, 2026, pp. 1-7.
Y. Xu, Y. Ni, T. He, R. Sun, Y. Jin and A. Zou, “Real-Time Scheduling and Analysis of Fixed-Priority Tasks on A Basic Heterogeneous Architecture with Multiple CPUs and Many PEs,” IEEE Transactions on Computers (TC), 2025, doi: 10.1109/TC.2025.3573602.
Y. Ni, Y. Xu, J. Chen, J. Li, C. Gill, X. Zhang, Y. Jin and A. Zou, “MATCH: Real-Time Scheduling of Multiple and Parallel Data Copies in Heterogeneous Architectures,” 2025 IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS), Irvine, U.S., 2025, pp. 349-361, doi: 10.1109/RTAS65571.2025.00040.
Y. Ni, T. Ma, J. Chen, C. Yang, S. Ye, Y. Xu, Y. Jin and A. Zou, “HARD: Hardening Real-Time Scheduling and Analysis for Accelerator Enabled Computing,” 2025 IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS), Irvine, U.S., 2025, pp. 389-401, doi: 10.1109/RTAS65571.2025.00043.
Y. Ni, J. Zhu, Y. Jin and A. Zou, “RTHeter: Simulating Real-Time Scheduling of Multiple Task in Heterogeneous Architectures,” 2025 Design, Automation & Test in Europe Conference & Exhibition (DATE), Lyon, France, 2025, pp. 1-7, doi: 10.23919/DATE64628.2025.10992806.
J. Chen, Y. Xu, Y. Ni, Y. Ma and A. Zou, “RICH: Heterogeneous Computing for Real-Time Intelligent Control Systems,” 2025 Design, Automation & Test in Europe Conference & Exhibition (DATE), Lyon, France, 2025, pp. 1-7, doi: 10.23919/DATE64628.2025.10993261.
R. Sun, Y. Ni, X. He, J. Zhao and A. Zou, "ONE-SA: Enabling Nonlinear Operations in Systolic Arrays For Efficient and Flexible Neural Network Inference," 2024 Design, Automation & Test in Europe Conference & Exhibition (DATE), Valencia, Spain, 2024, pp. 1-6, doi: 10.23919/DATE58400.2024.10546535.
Experience
-
Real-Time System Scheduling for Heterogeneous Systems
- Developed scheduling frameworks for accelerator-enabled computing platforms [RTAS'25]
- Investigating progress-aware real-time serving mechanisms [Under Review]
-
Mathematical Timing Analysis for Real-Time Computing
- Proposed response-time analysis methods for parallel data copies [RTAS'25]
- Working on improved analysis techniques for complex task models [Under Review]
-
Time Critical Computing in Modern Operating Systems
- Built simulation tools for heterogeneous architecture scheduling [DATE'25]
- Exploring latency-aware system optimization approaches [Under Review]
-
VE203 (MATH2030J): Discrete Mathematics
-
Topics: Logics, Relations, Advanced Counting, Group Theory, Number Theory, Graph Theory
-
VE280 (ECE2800J): Programming and Elementary Data Structures
-
Topics: Basic Linux, C++ Syntax, Testing, OOP, Dynamic Memory, STL
-
VG101 (ENGR1010J): Introduction to Computer and Programming
-
Topics: MatLab, C, C++ Syntax
-
VV186 (MATH1860J): Honors Mathematics
-
Topics: Function and Limits, Differentiation, Series, Integral of single variable
All the RC (recitation class) materials are available on github.
Please refer to my teaching page
-
Project Orca: Distributed Tensorflow and Pytorch
- Refactored github repo CI/CD workflow, eliminated unncessary dependencies, added daily/nightly testing
- Validated Yarn,K8S cluster setting, updated corresponding tutorials and documents;
- Created dockerfile for orca under different python environments
-
Project Fresian: High-Level Big-Data Recommendation;
- Developed Big Data recommendation example codes, created corresponding documents
- Released the corresponding tookit with docker-compose, k8s, helmchart
- Plan/ attend/ facilitate/ lead TA orientation, training workshops, seminars
- Work on the improvement of JI TA training programs, TA management, and other TA related issues
- Conduct spot check of TAs' recitation class, labs & office hours, provide constructive suggestions for their improvement, and file reports monthly for backup
Projects
RTHeter: A hierarchical simulator for Real-Time scheduling analysis on Heterogeneous architectures.
-
Key Features:
- 🔧 Supports ideal/practical models of CPUs, accelerators, and data copy engines
- 📊 Enables rapid evaluation of scheduling algorithms on customizable architectures
- 🤖 Facilitates RL-based scheduler discovery and architectural bottleneck analysis Technical Stack:
- C++ 20 simulator kernel with Python frontend interface
- Support for DAG tasks, self-suspension models, and priority-based scheduling
- Configurable simulation granularity from tick-level to millisecond scale Awards:
- 🥉 2025 ESWeek Embedded System Software Competition Second Runner-Up
- 📜 Published at DATE 2025 Conference Applications:
- Pre-silicon verification of real-time heterogeneous SoCs
- Design space exploration for CPU/accelerator resource allocation
- Benchmarking and comparison of scheduling policies
SAVE: A Vision-Based Bicycle Safety System for Alerting Collision with Vehicles.
-
Key Insights:
- 🚴Designed a cyclist equipment with safety embedded system.
- 🖥️Built a web server to allow real-time monitoring on both computers and mobiles.
- ⚡Improved program execution speed while extending battery life by scheduling. Dependencies:
- Raspberry Pi 4B, Intel NCS2, 2x 100 Binocular Camera, MPU6050
- C++ 17 on Linux, OpenVINO, OpenCV, Caddy Server 2 Awards:
- 🏆Silver Award of Undergraduate Thesis in JI 2023 Winter Expo
A recommendation system on Million Song Dataset
- Tools: Hadoop, Spark, Avro, Drill, Scikit-learn, LaTeX
- Retrieve the dataset, interprete on the data structure and turn into avro file in parall.
- Use drill (sqlite) to perform basic queries on dataset.
- Use both Spark and Mapreduce to recommend the most similar song with diverse customization.
- Reduce data dimension by PCA and effectively predict the year of the songs.
- Demostrate our idea by well typed beamer slides and an A1 poster.
Multi-threading optimization for simple database queries.
A simple shell (mumsh) implemented by system-level C.
- Key Points: system call, signal catching, pipe, built-in command.
- Support basic Unix commands w/o arguments, e.g.
ls -la. - Bash style file I/O redirection, e.g.
cat <1.txt > 2.txt. - Basic pipe, e.g.
echo 123 | grep 1. - Catch interrupt signal including
CTRL-DandCTRL-C. - Implement shell built-in command
pwdandcd.
Skills
Languages
C/C++
Shell Scripting
Python
MatLab
MySQL
Java
Libraries
NumPy
Pandas
OpenCV
scikit-learn
matplotlib
Frameworks
CUDA
Hadoop
Spark
Kubernates
PyTorch
Other
Git
Docker
LaTeX
Education
Shanghai, China
Degree: Bachelor of Engineer in ECE
Major: Electronic and Computer Engineering
Minor: Computer Science
GPA: 3.82/4.0
- ECE6703J Introduction to Computer-Aided Designs (A+)
- ECE4730J Advanced Embedded Systems (A)
- ECE4820J Introduction to Operating System (A+)
- ECE4720J Methods and Tools for Big Data (A+)
Core Courseworks Taken:
Shanghai, China
Degree: Master of Computer Science and Technology (in progress)
Thesis Proposal: Real-Time Scheduling on Heterogeneous Computing Architectures
GPA: 3.94/4.0
- Real-Time Scheduling on Heterogeneous Computing Systems
- Temporal Simulation and Task Modeling
- Time Critical Computing Support in Operating Systems
Research Topics: