Justin Zhao

California, United States
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts

Summary

🤩
Rockstar
Justin Zhao is a research engineer with 11 years of experience building and evaluating large-scale language models and ML systems, currently working on superintelligence safety evaluations at Meta and advising ML-driven product teams. He previously led core ML efforts at Predibase as a founding staff engineer and drove generative language model work at Google, shipping on-device text-generation models and large multilingual distillation pipelines. An active open-source contributor, he improved Ludwig’s data pipelines and text/image handling to make low-code model building more flexible. Based in New York, he blends deep systems and research expertise (C++/TensorFlow, deployment at scale) with product-minded engineering and mentorship. Outside engineering he co-founded Columbia’s premier pops orchestra, reflecting a long-standing parallel career in music that influences his collaborative and creative approach.
code11 years of coding experience
github-logo-circle

Github Skills (9)

pytorch10
machine-learning10
nlp10
python10
natural-language-processing10
text-processing10
modeling9
trainings9
feature-engineering9

Programming languages (3)

HTMLJupyter NotebookPython

Github contributions (5)

github-logo-circle
ludwig-ai/ludwig

Oct 2021 - Jan 2023

Low-code framework for building custom LLMs, neural networks, and other AI models
Role in this project:
userBack-end Developer & ML Engineer
Contributions:20 releases, 1472 reviews, 232 commits in 1 year 3 months
Contributions summary:Justin primarily focused on improving the `ludwig` library, which is a low-code framework for building custom LLMs and other AI models. Their contributions included implementing improvements to the binary feature processing pipeline, and adding features to the code to read image files. They also made significant changes to allow the library to handle text data with more flexibility, through updates to the string processing utilities. A large portion of the work was related to adding support for new functionality within the library's various modules.
ailow-codeneural-networkdeep-learningmachine-learning
machine-theory/lm-council

Jun 2024 - Jul 2025

LLMs sitting on a council together to decide, by consensus, who among them is the best.
Contributions:5 reviews, 22 PRs, 71 pushes in 1 year 1 month
consensus
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.
Request Free Trial