Gavin Ding
Member Of Technical Staff at Anthropic
San Francisco Bay Area United States
Summary
🤩
Rockstar🎓
Top SchoolGavin Ding is a senior researcher and engineer with 11 years of experience applying deep learning and reinforcement learning across finance, robotics, medical imaging, and geospatial mapping. Based in the San Francisco Bay Area, he has led and shipped production computer vision systems at Ecopia AI and guided RL research and teams at Borealis AI/RBC. His open-source work includes improving adversarial robustness tools (advertorch) and modernizing visualization utilities for convolutional nets, reflecting a mix of research rigor and pragmatic engineering. Gavin’s background in medical image analysis and robotics gives him a strong foundation in applied ML for safety-critical and perception-heavy domains. Currently focused on AI alignment at Anthropic, he brings both production experience and a research-first mindset to governance and robustness challenges. Colleagues describe him as someone who moves fluently between prototype research, test-driven improvements, and production deployments.
11 years of coding experience
15 years of employment as a software developer
Bachelor of Engineering Automation, Bachelor of Engineering Automation at Beihang UniversityMaster of Applied Science Medical Image Analysis, Master of Applied Science Medical Image Analysis at Simon Fraser University
Chinese, English
Stackoverflow
121reputation
2kreached
2answers
0questions
Github Skills (18)
data-visualizations10
pytorch10
adversarial-attacks10
python10
matplotlib10
machine-learning10
data-visualisation10
robustness10
robust10
data-visualization10
benchmark9
security9
benchmarking9
refactoring9
computer-vision8
Programming languages (3)
Github contributions (5)
BorealisAI/advertorch
Nov 2018 - Jun 2020
A Toolbox for Adversarial Robustness Research
Role in this project:
ML Engineer & Test Automation Engineer
ML Engineer & Test Automation EngineerContributions:156 commits, 65 PRs, 63 pushes in 1 year 6 months
Contributions summary:Gavin primarily focused on enhancing the robustness of the `advertorch` library, a toolbox for adversarial robustness research. Their contributions include adding comprehensive unit tests for utility functions, crucial for ensuring the reliability of the tools. They also corrected issues within existing attacks, such as fixing `veceps` in L2, and made improvements related to documentation. Furthermore, the user updated the repository by uploading the package to PyPI and creating the code for examples such as the tutorial on attacking an ImageNet model.
adversarial-learningpytorchadversarial-attacksbenchmarkingadversarial-perturbations
gwding/gwding.github.io
Dec 2015 - Jul 2022
Contributions:60 pushes in 6 years 7 months
jekyll-blogjekylljekyll-thememinutes
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.