Brandon Amos
New York, New York, United States
Summary
🤩
RockstarJing Dong is a Staff Software Engineer with a PhD from Georgia Tech and over a decade of experience building computer vision and robotics systems, currently leading ML-driven semantic HD mapping at Zoox in Redmond. Their work bridges research and production across SLAM, 3D reconstruction, and motion planning, with prior roles as a Staff Research Scientist and Tech Lead Manager at Meta. Jing has shipped low-level, differentiable optimization tooling to the community—contributing bug fixes and feature work to PyTorch solvers such as qpth and mpc.pytorch—demonstrating comfort from numerical linear algebra to system-level integration. Colleagues describe them as detail-oriented and pragmatic, able to turn research prototypes into vehicle-ready components while still contributing full‑stack improvements to research projects.
11 years of coding experience
Stackoverflow
940reputation
82kreached
8answers
10questions
Github Skills (25)
data-formats10
optimizations10
pytorch10
python10
formatting10
machine-learning10
data-format10
model-predictive-control10
mpc10
deep-learning10
quadratic-programming10
optimization10
linear-algebra9
bibtex9
markdown-it8
Programming languages (23)
Github contributions (5)
bamos/cv
Jun 2019 - Jan 2023
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:208 commits, 8 PRs, 432 pushes in 3 years 7 months
Contributions summary:Brandon primarily contributed to the project's front-end and back-end functionality, modifying the `generate.py` file extensively. Their contributions involved updating publication formatting, grouping publications by year, and fixing URL and spacing issues, demonstrating attention to detail. Furthermore, the user added features related to open-source repositories and GitHub stats, suggesting a focus on integrating external data into the project. These changes indicate an active role in maintaining and extending the project's core features.
locuslab/qpth
Jun 2019 - Sep 2019
A fast and differentiable QP solver for PyTorch.
Role in this project:ML Engineer
ML EngineerContributions:7 commits, 8 PRs, 146 pushes in 2 months
Contributions summary:Brandon primarily worked on updating and maintaining the `qpth` library, a differentiable QP solver for PyTorch. Their contributions focused on improving the core functionality of the solver, including fixing critical errors in the `lu_hack` function, updating the Function interface, and updating `lu_solve` calls. They also bumped the library version to reflect the changes, demonstrating involvement in project maintenance and release management.
pytorchdifferentiabledeep-learningoptimizationmachine-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.