Will Cromar
Software Engineer at Meta
Seattle, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolWill Cromar is a software engineer with a decade of experience building scalable machine learning infrastructure and TPU integrations, currently at Meta after five years at Google. He has contributed to high-profile open-source projects like pytorch/xla and multiple TensorFlow repositories, implementing core features (e.g., linspace on XLA) and enabling models to run efficiently on Cloud TPUs. Will’s work spans back-end systems, ML model maintenance, and cross-compatibility fixes that keep research code production-ready across TF1/TF2 and XLA environments. Based in Seattle and trained in computer science at the University of Central Florida, he blends practical engineering with deep familiarity of ML runtimes. Notably, his contributions often address non-obvious infrastructure gaps—static-shape support, TPUClusterResolver integration, and test-level fixes—that materially improve multi-TPU training reliability.
10 years of coding experience
6 years of employment as a software developer
Bachelor's degree, Computer Science, Bachelor's degree, Computer Science at University of Central Florida
Stackoverflow
199reputation
14kreached
8answers
0questions
Github Skills (31)
pytorch10
operation10
python10
tensorrt10
machine-learning10
tpu10
gcp10
keras10
deep-learning10
trainings10
tensorflow10
xla10
speech-recognition10
nlp10
tensor10
Programming languages (8)
Github contributions (5)
pytorch/xla
Feb 2022 - Jan 2023
Enabling PyTorch on XLA Devices (e.g. Google TPU)
Role in this project:
Back-end Developer
Back-end DeveloperContributions:924 reviews, 240 commits, 550 PRs in 11 months
Contributions summary:Will's contributions primarily focused on enhancing the `pytorch/xla` repository, which aims to enable PyTorch on XLA devices. Their work involved implementing and optimizing `torch.linspace()` functionality within the XLA environment. The user modified files related to XLA tensor operations, including changes to the aten_xla_type.cpp and linspace.cpp files, and added corresponding tests. These modifications involved integrating the new implementation and ensuring compatibility with XLA devices.
pytorchxladeep-learningtpucompiler
tensorflow/models
Oct 2019 - Oct 2020
Models and examples built with TensorFlow
Role in this project:ML Engineer
ML EngineerContributions:16 commits, 2 pushes, 8 branches in 1 year
Contributions summary:Will primarily contributed to the MNIST model within the TensorFlow models repository. Their work included adding and updating the 2.x version of the MNIST model, and fixing the test file. The user also made internal changes, including moving the mnist folder and modifying the resnet_ctl_imagenet_main.py file. These changes suggest a focus on model development and maintenance within the TensorFlow ecosystem.
deep-learningtensorflow
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.