2024 - Present
Harbin Institute of Technology, Shenzhen
B.Eng. candidate, Computer Science and Technology
Undergraduate student, ranked in the top 20% of the cohort. Shenzhen, China.
Quansheng GU
古权胜
I work where high-performance computing meets AI infrastructure. Across research and industry, I build systems for efficient inference.I work where high-performance computing meets AI infrastructure. Across research and industry, I build systems for efficient inference.
Current direction
AI infrastructure internships and systems research collaboration
Education
Research & Experience
May 2026 - Present
Research Intern
M³AIL Research GroupResearch intern with the M³AIL Research Group at Harbin Institute of Technology, Shenzhen.
- Building a systematic foundation in artificial intelligence algorithms.
- Collaborating with faculty and senior lab members on research projects.
Oct 2025 - Present
HPC Team Core Member
HITSZ-HPC / 南工九章Core member of the HITSZ supercomputing team, focusing on efficient inference for LLMs and world models.
- Competing in high-performance computing competitions, with responsibility for cluster design, deployment, and inference optimization.
- Serving as team captain, responsible for team operations and organization.
Awards
2026
Team captain
First Prize
ASC26 Final
ASC Student Supercomputer Challenge
2026
Team captain
Huawei ICT Competition, China
National Final, Third Prize
2024-2025
Individual
Harbin Institute of Technology, Shenzhen
Outstanding Student and Academic Scholarship
Skills
Experience across five compute ecosystems, from cluster software to distributed inference and runtime optimization.
| Ecosystem | NVIDIA GPU | AMD GPU | Huawei NPU | Huawei CPU | Hygon DCU |
|---|---|---|---|---|---|
| Architecture | Hopper / Ampere | RDNA 3 | AI Core | Armv9 / Armv8.2-A | DCU |
| Models | H100 / H20 / A100 | Radeon PRO W7900D | Ascend 910C / 910B | Kunpeng 920F / 920 | BW1000 |
| Software stack | TensorRT / CUDA Graphs | ROCm / HIP | CANN / Ascend C / HCCL | HPCKit / BiSheng | DTK / DAS |
Systems
Cluster and distributed computing
Linux, Distributed systems, MPI, Slurm, Shell, Git
AI Systems
Distributed and efficient model inference
LLM inference, DeepEP, Mooncake, PyTorch, ONNX
Performance
Profiling and bottleneck analysis
Nsight Systems, PyTorch Profiler, perf, msprof
Runtime Engineering
Deployment and execution optimization
TensorRT, CUDA Graphs, CANN
Outside work
Cycling, running, hiking, travel, photography, and music make up the other side of this site.
Open activity