Sam Foreman
Assistant Computational Scientist
Chicago, Illinois, United States
Summary
🤩
Rockstar🎓
Top SchoolSam Foreman is an Assistant Computational Scientist at Argonne National Laboratory with 11 years of experience applying AI/ML and high-performance computing to scientific problems. He specializes in large-scale model training and data pipeline engineering—demonstrated by contributions to Argonne’s ai-science training series where he implemented ImageNet preprocessing for large-scale computer vision workloads. His work bridges deep learning, HPC, and scientific simulation, from accelerating Hybrid Monte Carlo and Lattice QCD workflows to deploying TensorFlow models on many-core supercomputers like Theta and Aurora. Sam holds a Ph.D. in Physics and has a strong applied math and engineering physics background, enabling him to translate theoretical ideas (e.g., unsupervised learning for phase transitions) into production-scale training on exascale-class systems. Notably, he has developed generalized methods for neural-parameterized MCMC kernels that outperform traditional approaches, reflecting a blend of research depth and practical engineering. Based in Chicago, he focuses on training large models for science and scaling ML to run efficiently on next-generation HPC infrastructure.
11 years of coding experience
8 years of employment as a software developer
University of Illinois Urbana-ChampaignDoctor of Philosophy (Ph.D.), Physics, Doctor of Philosophy (Ph.D.), Physics at University of Iowa
c, python, English
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (7)
data-preprocessing10
computer-vision10
tensorflow10
python10
image-processing10
machine-learning9
horovod7
Programming languages (16)
Github contributions (5)
argonne-lcf/ai-science-training-series
Oct 2021 - Nov 2022
Role in this project:
ML Engineer
ML EngineerContributions:67 commits, 22 PRs, 32 pushes in 1 year 1 month
Contributions summary:Sam focused on implementing and updating the `ilsvrc_dataset.py` file within the `07_largeScaleTraining/src/ai4sci` directory. This file processes the ImageNet dataset for large-scale training, suggesting a focus on data loading and preprocessing for machine learning tasks. The changes include modifications to data loading, bounding box handling, and image resizing, demonstrating expertise in data pipeline development for computer vision.
saforem2/Megatron-DeepSpeed
Feb 2023 - Dec 2024
Ongoing research training transformer language models at scale, including: BERT & GPT-2
Contributions:2 reviews, 20 PRs, 59 pushes in 1 year 10 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.