Swanand Kadhe
Senior Research Scientist at IBM
San Francisco Bay Area United States
Summary
🤩
Rockstar🎓
Top SchoolSwanand Kadhe is a Senior Research Scientist in the San Francisco Bay Area specializing in large language models, data pipelines for pre- and post-training, and AI safety. With over a decade of interdisciplinary R&D across federated and distributed learning, blockchains, information theory and signal processing, he focuses on privacy-preserving ML algorithms and defenses against adversarial and trojan attacks. He has 50+ publications and multiple US patents, and contributes to influential open-source tooling—having added audio perturbation and backdoor demonstration capabilities to the widely used Adversarial Robustness Toolbox. Swanand’s work blends theoretical rigor from a PhD in Electrical and Communications Engineering with practical systems experience at IBM and top research labs, enabling resilient, fault-tolerant ML deployments. Colleagues rely on him to translate complex research into testable, production-ready modules that harden models in real-world, resource-constrained settings.
9 years of coding experience
12 years of employment as a software developer
Indian Institute of Technology KanpurDoctor of Philosophy (PhD), Electrical, Electronics and Communications Engineering, Doctor of Philosophy (PhD), Electrical, Electronics and Communications Engineering at Texas A&M UniversitySavitribai Phule Pune University
Github Skills (10)
machine-learning10
audio-processing10
adversarial-attacks10
librosa10
python10
numpy10
testing9
artificial-intelligence8
ai8
tensorflow8
Programming languages (2)
Github contributions (5)
Trusted-AI/adversarial-robustness-toolbox
Jun 2022 - Jun 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:3 reviews, 19 commits, 5 PRs in 12 days
Contributions summary:Swanand contributed to the Adversarial Robustness Toolbox by implementing and testing audio perturbation modules. They added functionality for inserting tone and audio triggers, which are crucial for creating adversarial examples in audio. These additions included writing the core Python code for the perturbations and developing unit tests to validate the functionality. The user also created a notebook demonstrating an audio backdoor attack.
extractionpythonfairness-mlrobustnessadversarial-machine-learning
Swanand-Kadhe/adversarial-robustness-toolbox
Jun 2022 - Mar 2023
Adversarial Robustness Toolbox (ART) - Python Library for Machine Learning Security - Evasion, Poisoning, Extraction, Inference - Red and Blue Teams
Contributions:16 pushes, 5 branches in 9 months
extractionpythonrobustnessadversarial-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.