Archie Gertsman
Software Engineer at Caterpillar Inc.
Chicago, Illinois, United States
Summary
👤
Senior🎓
Top SchoolArchie Gertsman is a software engineer with nine years of experience building production ML systems and developer infrastructure, currently focused on MLOps for serving and monitoring models at Caterpillar in Chicago. He combines an academic grounding in math and industrial engineering (MS, UIUC, 3.94 GPA) with hands-on work accelerating Spark scheduling as a graduate researcher and shipping ML/data tooling in industry internships. Archie contributes to PyTorch Geometric—adding and benchmarking a Learnable Commutative Monoid aggregation—demonstrating a knack for designing and vectorizing novel graph neural network components. Pragmatic and research-minded, he bridges model innovation and reliable deployment, often surfacing subtle algorithmic improvements that scale in practice.
9 years of coding experience
2 years of employment as a software developer
Master of Science - MS, Industrial Engineering, 3.94, Master of Science - MS, Industrial Engineering, 3.94 at University of Illinois Urbana-Champaign
Russian, English
Stackoverflow
1,631reputation
92kreached
21answers
80questions
Github Skills (16)
algorithm10
data-structures10
geometric-deep-learning10
algorithms10
pytorch10
machine-learning10
deep-learning10
graph-neural-network10
graph-convolutional-networks10
python10
data-structure10
vector6
ios6
struct6
swift6
Programming languages (3)
Github contributions (5)
pyg-team/pytorch_geometric
Jun 2022 - Dec 2024
Graph Neural Network Library for PyTorch
Role in this project:
ML Engineer
ML EngineerContributions:1 review, 8 PRs, 9 comments in 2 years 6 months
Contributions summary:Archie primarily contributed to the `LCMAggregation` module within the PyTorch Geometric library. Their work involved implementing, optimizing, and testing the Learnable Commutative Monoid aggregation, including vectorization for improved performance. They also developed and integrated a benchmark task based on the second minimum problem to showcase the effectiveness of `LCMAggregation`, along with adding random permutation transformations. This demonstrates a focus on developing and benchmarking novel graph neural network components.
pytorchgraph-convolutional-networksgeometric-deep-learningdeep-learningneural-graph
ArchieGertsman/spark-sched-sim
Feb 2022 - Mar 2023
A Gymnasium environment for simulating job scheduling in Apache Spark
Contributions:111 commits, 4 PRs, 129 pushes in 1 year 1 month
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.