Andrew Or
Research Scientist at Meta
New York City Metropolitan Area United States
Summary
🤩
Rockstar🎓
Top SchoolAndrew Or is a Research Scientist with 13 years of experience building and optimizing distributed systems and machine learning tooling, currently based in the New York City area. He holds a PhD from Princeton where his thesis abstracted systems challenges in distributed deep learning, and has applied that expertise at Meta, Databricks, and Microsoft research projects. A prolific open-source contributor and Apache Spark PMC member, he has improved core Spark data structures and performance tooling while also contributing quantization and export features across major PyTorch repositories. He codes primarily in Python, Scala, Java and C++, and brings both low-level systems engineering and ML model deployment experience—unusually pairing memory- and performance-focused Spark work with deep involvement in PyTorch quantization and QAT. His background suggests a knack for finding pragmatic system-level fixes that unlock practical ML scalability.
13 years of coding experience
8 years of employment as a software developer
B.S., Electrical Engineering and Computer Science, B.S., Electrical Engineering and Computer Science at University of California, BerkeleyDoctor of Philosophy (Ph.D.), Computer Science, Doctor of Philosophy (Ph.D.), Computer Science at Princeton University
Github Skills (24)
apache-spark10
pytorch10
python10
memory-management10
big-data10
machine-learning10
data-structure10
scala10
performance-testing10
performance-optimization10
lora10
data-structures10
quantization10
technical-writing9
java9
Programming languages (6)
Github contributions (5)
pytorch/pytorch
Dec 2021 - Jan 2023
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Role in this project:
ML Engineer
ML EngineerContributions:1 release, 506 reviews, 423 commits in 1 year 1 month
Contributions summary:Andrew's commits primarily involve modifications to the PyTorch framework, specifically related to the implementation and testing of new features and APIs for quantized models. Contributions include refactoring existing quantization code for better type safety and user experience, adding support for new patterns in the PT2E (PyTorch 2.0 Export) quantization flow, and fixing numerics. The user also addressed several edge cases and issues that arose in the existing quantization framework, and also made improvements to the internal representation of the code.
pythongpu-accelerationdeep-learninggpunumpy
apache/spark
Nov 2014 - Nov 2014
Apache Spark - A unified analytics engine for large-scale data processing
Role in this project:Back-end Developer
Back-end DeveloperContributions:1 commit, 258 PRs, 7026 comments in 1 day
Contributions summary:Andrew primarily contributed to refactoring and optimization of core data structures within the Apache Spark project, specifically focusing on improving the `ExternalAppendOnlyMap`. The commits involve moving code to utility classes, implementing features, and optimizing memory usage. The changes include improvements to memory handling and sorting, aimed at improving performance.
analyticspythondata-processingsqlapache
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.