Graham Neubig
Co-Founder at OpenHands
Pittsburgh, Pennsylvania, United States
Summary
🤩
Rockstar🎓
Top SchoolGraham Neubig is an associate professor of computer science at Carnegie Mellon University, co-founder of OpenHands, and Chief Scientist at All Hands AI, with 15 years of experience at the intersection of AI, machine learning, and natural language processing. He blends deep academic research—PhD and master’s training from Kyoto University—with hands-on engineering, having contributed core optimizations to DyNet and authored practical code for neural NLP models used in teaching and research. His work emphasizes openness and accessibility: open publishing of papers, course materials, video lectures, and software to lower barriers to advanced NLP. He has founded and led startups, bridging research to product during his time as CEO of Inspired Cognition. Based in Pittsburgh, he brings a global background from education in the U.S. and Japan and a demonstrated focus on efficient core ML implementations rather than only high-level prototypes. Colleagues know him for shipping reproducible, educational resources that accelerate both research and real-world NLP adoption.
15 years of coding experience
10 years of employment as a software developer
Doctor of Philosophy - PhD Informatics, Doctor of Philosophy - PhD Informatics at Kyoto UniversityBachelor's degree Computer Science, Bachelor's degree Computer Science at University of Illinois Urbana-Champaign
Github Skills (19)
pytorch10
c-language10
operation10
python10
tensorrt10
word-embeddings10
machine-learning10
tensorflow10
natural-language-processing10
language-modeling10
neural-network10
nlp10
tensor10
cprogramming-language10
linear-algebra9
Programming languages (18)
Github contributions (5)
clab/dynet
May 2015 - Mar 2020
DyNet: The Dynamic Neural Network Toolkit
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:8 releases, 1 review, 1052 commits in 4 years 10 months
Contributions summary:Graham made several commits focused on enhancing the functionality of the DyNet toolkit. Specifically, they worked on making nodes use the default device, fixing warnings, and "tensorifying" component-wise operations. Their contributions indicate a focus on optimizing the core operations and making them more efficient, which is vital in a numerical deep learning library.
dynetdynamic-neural-networkdeep-learningneural-networksmachine-learning
neubig/nn4nlp-code
Aug 2017 - Jan 2020
Code Samples from Neural Networks for NLP
Role in this project:ML Engineer
ML EngineerContributions:48 commits, 22 PRs, 41 pushes in 2 years 5 months
Contributions summary:Graham made modifications to several code samples related to neural networks for NLP, indicating a focus on machine learning within this repository. They implemented and modified code for various models, including bag-of-words, continuous bag-of-words (CBOW), and deep CBOW architectures, showcasing an involvement in core model development. The user also added new code, specifically for language modeling, including both neural network and log-linear models. Furthermore, they added improvements related to efficiency, negative sampling, and binary prediction.
nlpnatural-language-processingneural-networksmachine-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.