Benjamin Loison
PHD Student at AMIAD - Agence Ministérielle pour l'IA de Défense
Versailles, Ile-de-France
Summary
👤
Senior🎓
Top SchoolBenjamin Loison is a PhD student and researcher in image forensics and theoretical computer science, specializing in detection of deepfakes through analysis of image noise patterns and PRNU on modern smartphones. Trained at École Normale Supérieure Paris-Saclay after Classe Préparatoire, he combines a strong mathematical background with ten years of hands-on experience across research internships and engineering roles at institutions like DFINITY, Inria, IMT Atlantique and CEA. His work spans blockchain protocol design and practical tooling (contributing front-end improvements to the high-profile Internet Computer developer portal) as well as deep-learning notebook fixes and forensic methods, reflecting fluency from low-level systems to applied ML. Curious and meticulous, he is drawn to decentralization and robustness, seeking theoretical foundations that inform practical, reproducible defenses against manipulated media.
10 years of coding experience
Research Year in Artificial Intelligence, Artificial Intelligence, Research Year in Artificial Intelligence, Artificial Intelligence at École normale supérieure Paris-SaclayClasse Préparatoire aux Grandes Écoles MPSI/MP*, Classe Préparatoire aux Grandes Écoles MPSI/MP* at Lycée Fénelon
German, French, English
Stackoverflow
4,727reputation
150kreached
248answers
5questions
youtube-data-api
top-1%
google-api
top-5%
youtube
top-1%
youtube-api
top-1%
Github Skills (29)
pytorch10
learn-ruby-on-rails10
python10
youtube-api10
user-manual10
user-interface10
basics10
youtube10
deeplearning-ai10
deep-learning10
youtube-data-api10
react10
google-api9
javascript9
typescript9
Programming languages (32)
Github contributions (5)
dfinity/portal
Oct 2022 - Oct 2022
Internet Computer Developer Portal
Role in this project:
Front-end Developer
Front-end DeveloperContributions:6 commits, 2 PRs in 11 days
Contributions summary:Benjamin primarily contributed to the front-end of the Dfinity portal, focusing on user interface improvements and content updates. They corrected broken links in Bitcoin integration tutorials, ensuring users can easily access relevant educational material. The user also updated various components, including the roadmap and featured articles, enhancing the overall user experience and navigation of the platform. Significant changes focused on the videos and web speed pages.
web-appweb-applicationdockerportalinternet
dataflowr/notebooks
Sep 2022 - Sep 2022
code for deep learning courses
Role in this project:Data Scientist
Data ScientistContributions:7 commits, 1 PR in 1 day
Contributions summary:Benjamin primarily focused on correcting typos and making minor adjustments within the provided notebooks, specifically within the context of deep learning courses. They corrected typos across several notebooks, demonstrating a good understanding of the project's content and structure. Furthermore, the user made targeted code adjustments in the notebooks related to topics like Variational Autoencoders (VAEs) and Federated Poisoning. They also addressed issues related to variable definitions, showing a focus on detail and accuracy within the code examples.
deep-learningpytorchmachine-learning
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.