Andrew Ilyas
Summary
🤩
RockstarAndrew Ilyas is a machine learning engineer with 11 years of experience and active PhD-level research background from MIT focused on robust ML. He blends research rigor with production-oriented engineering, contributing core improvements to MadryLab’s widely used robustness library—adding dataset support, training refinements, and documentation for adversarially robust ImageNet models. Comfortable across the stack, he has also shipped user-facing features as a full-stack developer on the Falcon Chrome extension, improving text extraction, UI preferences, and styling. His work sits at the intersection of adversarial robustness research and practical tooling, making state-of-the-art models more reproducible and usable. Colleagues rely on him to translate complex research ideas into reliable code and demonstrable artifacts like model checkpoints and extension features.
11 years of coding experience
Github Skills (17)
pytorch10
javascript10
python10
imagenet10
browser-extension10
machine-learning10
chrome-extension10
chrome-app10
deep-learning10
robustness10
neural-network10
chrome-plugin10
css9
html9
datasets8
Programming languages (8)
Github contributions (5)
MadryLab/robustness
Aug 2019 - Nov 2021
A library for experimenting with, training and evaluating neural networks, with a focus on adversarial robustness.
Role in this project:
ML Engineer
ML EngineerContributions:7 releases, 1 review, 73 commits in 2 years 3 months
Contributions summary:Andrew primarily contributes to the development and improvement of machine learning models within the robustness library. Their work includes linking to model checkpoints, adding support for random starts in training, and introducing a custom dataset for ImageNet. Further contributions involve changing default training parameters for ImageNet and documenting the use of the library and its pretrained models.
neural-network
lengstrom/falcon
Jul 2016 - Oct 2016
Chrome extension for full text history search!
Role in this project:Full-stack Developer
Full-stack DeveloperContributions:1 release, 35 commits, 10 PRs in 3 months
Contributions summary:Andrew primarily focused on developing and refining the Falcon Chrome extension. Their contributions include adding features for relevant text extraction, including the addition of title to relevant text. They also added history display and clear functionality to preferences, and made styling improvements. These commits demonstrate a focus on both frontend and backend aspects of the extension.
chrome-extension
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.