Yaohui Liu
软件工程师 at UST
Portland, Oregon, United States
Summary
🤩
RockstarYaohui Liu is a software engineer based in Portland, Oregon with 7 years of experience focused on AI infrastructure and distributed training systems. Currently at UST, he brings practical backend engineering chops to large-scale ML toolchains and production inference stacks. An active open-source contributor, Yaohui has extended TensorFlow’s C API and added front-end and operator support in OpenVINO, demonstrating both low-level systems fluency and ML framework interoperability. He favors robustness and test-driven improvements—adding unit tests, fixing plugin exceptions, and optimizing logging across bindings. Comfortable working across C/C++ and Python ecosystems, he bridges researcher-facing APIs and production deployment needs. Notably, his contributions improve accessibility for C/C++ developers using major ML frameworks, a niche often overlooked in AI infra work.
7 years of coding experience
Github Skills (14)
pytorch10
machine-learning10
c-language10
deep-learning10
inference10
api10
tensorflow10
cprogramming-language10
python10
apidoc10
ai10
openvino10
computer-vision9
neural-network8
Programming languages (11)
Github contributions (5)
openvinotoolkit/openvino
Jan 2023 - Jul 2024
OpenVINO™ is an open source toolkit for optimizing and deploying AI inference
Role in this project:
ML Engineer
ML EngineerContributions:25 reviews, 11 PRs, 57 comments in 1 year 5 months
Contributions summary:Yaohui contributed to the OpenVINO toolkit by implementing complex type support for the rank operation in the TensorFlow frontend, adding tests for the new functionality. They also fixed an issue related to unexpected exception information within the BATCH plugin. Furthermore, the user optimized logging in the PyTorch Dynamo bindings and added support for the `ones_like` and `zeros` operations in the FX front-end, along with supporting the `aten::_unique2` operation and adding unit tests for it. The user also supported the `aten::index_copy_` operation.
inference-enginepytorchmodel-optimizerdeep-learninggpu
tensorflow/tensorflow
Oct 2020 - Jun 2023
An Open Source Machine Learning Framework for Everyone
Role in this project:Back-end Developer
Back-end DeveloperContributions:1 review, 1 PR, 9 comments in 2 years 8 months
Contributions summary:Yaohui primarily contributed to the TensorFlow C API, adding new APIs that correspond to existing Python C APIs. Their work involved modifying header and source files to incorporate these new functionalities, as well as adding helper functions. Further contributions fixed errors related to the C API, indicating a focus on maintaining and expanding the C interface for TensorFlow. These changes likely aim to improve the usability and accessibility of TensorFlow for C/C++ developers.
pythondata-sciencedeep-learningmlmachine-learning
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.