Cong Xu
Applied Scientist at Amazon
San Francisco, California, United States
Summary
👤
Senior🎓
Top SchoolCong Xu is an Applied Scientist in San Francisco with nine years of experience building high-performance distributed ML and HPC systems, currently working at Amazon after a deep-learning software engineering tenure at Intel. He combines PhD-level research in parallel I/O and MPI with hands-on expertise in distributed LLM training (Megatron, FSDP, DeepSpeed), NCCL/RDMA optimizations, and cluster administration for GPU and Infiniband environments. Proficient in C/C++, Python and Java, he bridges systems and ML stacks—from Lustre/MPI-IO and HDF5 to PyTorch/TensorFlow—enabling scalable training and big-data analytics pipelines. His background includes published research on scalable MPI AlltoallV algorithms and practical experience shipping MapReduce/Hive projects and full-stack data collection tools. Colleagues rely on him to translate low-level performance engineering into robust, production-ready ML workflows.
9 years of coding experience
11 years of employment as a software developer
Bachelor's degree, Computer Science, Bachelor's degree, Computer Science at Beijing University of Post and TelecommunicationsMaster's degree, Computer Science, Master's degree, Computer Science at Auburn University
English, Chinese
Github Skills (44)
python10
data-science10
machine-learning10
ml10
deep-learning10
tensorflow10
deep-neural-networks10
neural-network10
model-compression9
nas9
hyperparameter-tuning8
neural-architecture-search8
automated-machine-learning8
storage8
feature-engineering8
Programming languages (3)
Github contributions (5)
congxu-ml/nni
Mar 2022 - Mar 2022
An open source AutoML toolkit for automate machine learning lifecycle, including feature engineering, neural architecture search, model compression and hyper-parameter tuning.
Contributions:4 pushes, 1 branch in 7 days
pythonneural-architecture-searchengineeringtensorflowhyper
congxu-ml/dawn-bench-entries
Oct 2018 - Oct 2018
Contributions:1 push in 1 day
pytorchaprilend-to-enddeep-learningbenchmark
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.