Corentin Jemine
Machine Learning Engineer at Sanas
Liège, Liège, Belgium
Summary
🤩
Rockstar🎓
Top School
Top expert inDeep Learning and Computer Vision Technologies
Corentin Jemine is a machine learning engineer with 11 years of experience specializing in streaming inference and speech technologies, currently based in Liège and available for remote contract work. He has driven production-grade ML systems from data collection to low-latency inference at companies like Resemble AI and MetaVoice, and now at Sanas. Corentin built TorchStream as a solo open-source project for streaming DNN inference and contributed to the popular Real-Time-Voice-Cloning repo, focusing on feature extraction and VAD—underscoring his strength in data processing for audio models. His background spans foundational MLE and SWE work, including large-scale data ops for terabyte-scale voice model training and modular, maintainable inference engines. He combines deep CS training from the University of Liège with hands-on engineering to bridge research and production for real-time, performance-critical ML. Colleagues rely on him for pragmatic pipelines that enable streaming, inpainting, and sequential model extensions in applied speech systems.
11 years of coding experience
1 year of employment as a software developer
Master's degree Data Science/Machine Learning, Master's degree Data Science/Machine Learning at University of Liège
English, French, Dutch
Github Skills (6)
data-preprocessing10
pytorch10
deep-learning10
speech-processing10
python10
feature-extraction10
Programming languages (13)
Github contributions (5)
CorentinJ/Real-Time-Voice-Cloning
Sep 2018 - Sep 2022
Clone a voice in 5 seconds to generate arbitrary speech in real-time
Role in this project:
ML Engineer
ML EngineerContributions:16 reviews, 249 commits, 55 PRs in 4 years
Contributions summary:Corentin contributed to the feature extraction process, demonstrating a focus on data processing within the deep learning domain. The user modified and added code within the context of a voice cloning project, including data loading and processing and the implementation of a VAD algorithm. This user's activities suggest a strong focus on machine learning model development and data preparation.
speechdeep-learningpytorchtensorflowtts
jplhughes/voice_cloning
May 2019 - Nov 2019
Clone and transcribe your voice in 5 seconds and generate a synthesised continuation with GPT-2. Many new features compared to forked repo.
Contributions:97 commits in 5 months
gpt
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.