Jin Zhang
Software Engineer at Harvey
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top SchoolJin Zhang is a software engineer with a rare blend of academic physics rigor and production-grade ML and systems engineering, currently building LLM-driven products in Mountain View. Over a decade-plus career spanning YouTube, Loon, Databricks, and now Harvey, Jin has led teams and technical efforts that move research into scalable services—from large-scale recommendation and personalization at YouTube to fleet simulation and network optimization at Loon. At Databricks he was a tech lead for MLflow and MLOps systems that serve thousands of data scientists and support a widely used open-source project; his OSS contributions include improving TensorBoard autologging and artifact capture in mlflow/mlflow. He is comfortable across the stack: model training and serving, distributed simulation, and production reliability and compliance for enterprise model endpoints. Trained as a condensed matter theorist (PhD, Warwick) with a postdoctoral record in computational quantum chemistry, he brings strong analytical instincts and a history of turning novel research ideas into practical, high-throughput platforms. Colleagues rely on him for pragmatic design, measurable cost savings, and bridging complex technical domains into delivered products.
5 years of coding experience
14 years of employment as a software developer
M.Sc Condensed matter physics, M.Sc Condensed matter physics at Peking UniversityPh.D Condensed matter theory, Ph.D Condensed matter theory at University of Warwick
Chinese, English
Github Skills (8)
machine-learning10
logging10
mlflow10
tensorflow10
python10
testing9
ai9
model-management9
Programming languages (1)
Github contributions (5)
mlflow/mlflow
Feb 2021 - Nov 2022
Open source platform for the machine learning lifecycle
Role in this project:
ML Engineer
ML EngineerContributions:542 reviews, 51 commits, 85 PRs in 1 year 9 months
Contributions summary:Jin primarily contributed to the implementation of Tensorboard logging within the autologging functionality for TensorFlow Estimator APIs in the `mlflow/mlflow` repository. These changes included capturing Tensorboard event files as artifacts and ensuring the correct logging of metrics. The user also made changes to the tests to verify the logging of tensorboard artifacts. Furthermore, the user added retries and updated the timeout configuration.
pythonlifecyclemlmachine-learningincremental-learning
jinzhang21/mlflow
Nov 2021 - Jun 2023
Open source platform for the machine learning lifecycle
Contributions:53 pushes, 13 branches in 1 year 7 months
pythonmachine-learninglifecyclemachine-learning-lifecycle
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.