Qiaoling Chen
ResearchAsistant at OneFlow
Singapore, Singapore
Summary
🤩
Rockstar🎓
Top SchoolQiaoling Chen is a machine learning systems researcher and engineer with three years of experience optimizing large-scale distributed training and production AI frameworks. Based at NTU and contributing to OneFlow, she has driven up to 200% cluster performance gains by identifying and filling execution "bubbles" during Megatron-DeepSpeed GPT training and automating small-model insertion without impacting main-model throughput. Her work spans hands-on profiling with NVIDIA tools (nsys, DCGM), CUDA/PyTorch/NCCL tuning, and C++ operator development, delivering measurable improvements in memory and throughput across RTX 3090 and A100 deployments. Beyond performance tuning she has implemented OneFlow OCR, AI-writer, and BERT-NER solutions and refactored framework documentation to align with PyTorch APIs, reflecting both deep technical skill and attention to developer experience.
3 years of coding experience
Master of Technology - MTech, Enterprise Business Analysis, Master of Technology - MTech, Enterprise Business Analysis at 新加坡国立大学
Github Skills (11)
nlp7
transformers7
scale7
transformer-models6
pytorch6
machine-learning6
large-language-models6
bert6
deep-learning5
transformer5
language-model4
Programming languages (2)
Github contributions (5)
ChenQiaoling00/DeepSpeed-AMSP
Aug 2024 - Mar 2025
DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
Contributions:1 PR, 4 pushes, 1 branch in 7 months
ChenQiaoling00/ChenQiaoling00.github.io
Jun 2024 - Mar 2025
AcadHomepage: A Modern and Responsive Academic Personal Homepage
Contributions:22 pushes in 9 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.