Ambrish Rawat
Senior Research Scientist, AI Foundations
Town of Yorktown, New York, United States
Summary
🤩
Rockstar🎓
Top SchoolAmbrish Rawat is a Senior Research Scientist in AI Foundations with 11 years of experience translating adversarial-robustness research into production-ready safeguards for large-scale AI deployments. At IBM he has led cross-functional programs around red-teaming generative AI, building guardrails for watsonX and hardening open-source foundation models like Granite through tooling such as Granite Guardian. His work spans adversarial ML, federated learning, privacy-enhancing technologies, and Bayesian methods, with publications at top conferences and multiple granted U.S. patents. An active contributor to the Adversarial Robustness Toolbox, he has implemented practical robustness metrics and attention-map features that bridge research evaluation and applied security. Trained at IIT Delhi and Cambridge, he combines rigorous academic foundations with a track record of delivering operationally reliable AI in global research and product settings.
11 years of coding experience
10 years of employment as a software developer
Indian Institute of Technology Delhi (IIT Delhi)Master of Philosophy (M.Phil.), Machine Learning and Machine Intelligence, Master of Philosophy (M.Phil.), Machine Learning and Machine Intelligence at University of Cambridge
English, Hindi
Github Skills (5)
keras10
machine-learning10
adversarial-attacks10
tensorflow10
python10
Programming languages (3)
Github contributions (5)
Trusted-AI/adversarial-robustness-toolbox
Jul 2017 - Mar 2022
Adversarial Robustness Toolbox (ART) - Python Library for Machine Learning Security - Evasion, Poisoning, Extraction, Inference - Red and Blue Teams
Role in this project:
ML Engineer
ML EngineerContributions:4 reviews, 95 commits, 4 PRs in 4 years 9 months
Contributions summary:Ambrish's commits focus on implementing and integrating machine learning metrics within the Adversarial Robustness Toolbox. Their contributions involve adding and modifying code in `src/metrics.py` and `src/metrics_unittest.py` to incorporate metrics such as MMD (Maximum Mean Discrepancy), and nearest-neighbor distance, which are relevant for evaluating the robustness of machine learning models against adversarial attacks. The user also implemented a new feature: the attention map. In addition the user added a new class for K Nearest Neighbors
extractionpythonfairness-mlrobustnessadversarial-machine-learning
ambrishrawat/adversarial-robustness-toolbox
Apr 2020 - Nov 2021
Python library for adversarial machine learning (evasion, extraction, poisoning, verification, certification) with attacks and defences for neural networks, logistic regression, decision trees, SVM, gradient boosted trees, Gaussian processes and more with multiple framework support
Contributions:131 pushes, 5 branches in 1 year 7 months
boosted-treesextractionpythonadversarial-machine-learningadversarial
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.