Adelin T
Japan
Summary
🤩
Rockstar🎓
Top SchoolAdelin T is an expert in machine learning assurance and security with eight years of experience bridging adversarial ML research and applied cybersecurity across industry and standards bodies. Currently holding leadership roles at ISO/IEC AI and Woven by Toyota, he brings hands-on principal engineering experience from Trail of Bits and consulting pedigree from Verizon and EY. His open-source contributions to the well-known CleverHans adversarial library show deep familiarity with attack algorithms and TensorFlow internals, highlighting practical skills in threat modeling for ML systems. Trained at Oxford (MSc, Distinction) and with PhD work in ML security at the University of Toronto, he blends academic rigor with pragmatic engineering. He has a track record of moving research into production-grade assurance processes for large enterprises and standards organizations. Notably, he has transitioned from traditional pentesting and red-team consulting into shaping ML-specific security practices at both corporate and international standard levels.
8 years of coding experience
3 years of employment as a software developer
Master of Science (MSc), Computer Science, Distinction, Master of Science (MSc), Computer Science, Distinction at University of OxfordDoctor of Philosophy - PhD, Machine Learning and Security, On hold, Doctor of Philosophy - PhD, Machine Learning and Security, On hold at University of TorontoDiplôme d'ingénieur, Informatique, Diplôme d'ingénieur, Informatique at Telecom ParisTechLycée charlemagneCollège StanislasLycée Louis le Grand
French, English, Japanese, German, Chinese
Github Skills (8)
machine-learning10
adversarial-attacks10
tensorflow10
python10
security10
benchmarking9
benchmark9
numpy8
Programming languages (3)
Github contributions (5)
cleverhans-lab/cleverhans
Jan 2021 - Sep 2021
An adversarial example library for constructing attacks, building defenses, and benchmarking both
Role in this project:
ML Engineer
ML EngineerContributions:1 review, 15 commits, 10 PRs in 8 months
Contributions summary:Adelin made several updates to the `fast_gradient_method.py` and `projected_gradient_descent.py` files, suggesting a focus on adversarial attack algorithms. These updates modified the loss function definitions, and added parameters like targeted, rand_init and rand_minmax. The user also updated `utils_tf.py`, likely to incorporate changes related to the TensorFlow framework. The contributions suggest involvement in enhancing and refining existing attack methods within the CleverHans framework.
benchmarkingrobustnessadversarial-machine-learningsecurityadversarial
alkaet/alkaet.github.io
Apr 2022 - Dec 2024
Contributions:21 pushes in 2 years 8 months
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.