Wei Ho
Software Engineer at Meta
San Francisco, California, United States
Summary
👤
Senior🎓
Top SchoolWei Ho is an experienced software engineer in applied machine learning with a decade of industry experience building production-grade language and translation systems at Meta and earlier ML infrastructure and serving platforms at Google. He combines deep practical knowledge of neural machine translation with backend engineering skills, contributing to prominent open-source projects like fairseq and PyTorch Translate where he improved dataset/ checkpointing, model exporting, and memory efficiency. Based in San Francisco, he has a track record of moving models from research into deployable formats (including ONNX export) and optimizing tooling that teams rely on. A Princeton summa cum laude CS graduate, he blends academic rigor with hands-on systems work and a penchant for refactoring messy code into reliable, maintainable components.
10 years of coding experience
8 years of employment as a software developer
B.S.E., Computer Science, summa cum laude, B.S.E., Computer Science, summa cum laude at Princeton University
Github Skills (22)
pytorch10
artificial-intelligence10
checkpoint10
python10
machine-learning-models10
machine-learning10
data-export10
onnx10
exporter10
checkpointing10
natural-language-processing10
nlp10
exports10
data-set9
code-optimization9
Programming languages (2)
Github contributions (5)
pytorch/translate
May 2018 - Jul 2020
Translate - a PyTorch Language Library
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:67 commits, 39 PRs, 14 pushes in 2 years 2 months
Contributions summary:Wei primarily contributed to the PyTorch Translate library by addressing installation issues, improving code readability through the fixing of f-string issues, and refactoring and refactoring code to improve memory usage and efficiency. They also worked on refining example scripts for training, generating, and exporting models. Furthermore, they made changes related to model exporting and loading, specifically for ONNX component export, demonstrating familiarity with model conversion and deployment.
pytorchartificial-intelligencemachine-learningonnx
facebookresearch/fairseq
May 2018 - Jun 2022
Facebook AI Research Sequence-to-Sequence Toolkit written in Python.
Role in this project:Backend Developer
Backend DeveloperContributions:46 commits, 3 PRs, 3 comments in 4 years 1 month
Contributions summary:Wei primarily contributed to the `fairseq` library, focusing on improving the dataset and checkpointing functionalities. Their commits include updating dataset code for compatibility with other projects, adding options to load checkpoints, fixing print statements, and refactoring checkpoint-related utilities. Furthermore, the user addressed various issues and incorporated code formatting changes within the project.
aipythonsequence-to-sequencepytorchartificial-intelligence
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.