Guillaume Klein
Machine Learning Engineer at Apple
Paris, Ile-de-France
Summary
🤩
Rockstar🎓
Top SchoolGuillaume Klein is a Senior Staff Machine Learning Engineer with 11 years of experience, currently optimizing large language model serving and inference at Apple. He authored and led the OpenNMT ecosystem and created CTranslate2, a widely used high-performance inference engine that made Transformer models cheaper and more practical to run in production. His background spans research and engineering—building training and serving backends, speech recognition servers, and low-level C++ optimizations for numerical stability and vectorized performance. An active open-source maintainer, he has contributed to TensorFlow Addons and numerous ML repos, balancing API design, compatibility fixes, and performance engineering. Based in Paris, he combines deep NLP research roots with hands-on systems work that bridges model development and efficient deployment.
11 years of coding experience
8 years of employment as a software developer
Engineer's degree, Computer Engineering, Engineer's degree, Computer Engineering at EPITA: Ecole d'Ingénieurs en Informatique
French, English, German
Stackoverflow
504reputation
17kreached
8answers
8questions
Github Skills (42)
algorithm10
code-optimization10
speech-to-text10
pytorch10
algorithms10
machine-translation10
c-language10
language-model10
neural-machine-translation10
python10
attention-mechanism10
whisper10
torch-hub10
machine-learning10
inference10
Programming languages (12)
Github contributions (5)
OpenNMT/CTranslate2
Apr 2018 - Jan 2023
Fast inference engine for Transformer models
Role in this project:
Back-end Developer
Back-end DeveloperContributions:101 releases, 68 reviews, 1849 commits in 4 years 10 months
Contributions summary:Guillaume made contributions that removed and optimized the implementation of operations within the code. Specifically, the user was responsible for removing unused functions and enhancing existing ones for efficiency. The user's work involved vectorization techniques and applying specific C++ libraries for more precise numerical computations.
inference-engineavxbertintrinsicsopenmp
SYSTRAN/faster-whisper
Feb 2023 - Mar 2023
Faster Whisper transcription with CTranslate2
Role in this project:ML Engineer
ML EngineerContributions:7 releases, 33 reviews, 56 commits in 1 month
Contributions summary:Guillaume primarily contributed to the `faster-whisper` project by implementing and refining features related to the Whisper model's transcription capabilities. Their work includes adding functionalities such as initial prompt and prefix parameters, as well as parameters to control word timestamps and suppress certain tokens. The user also improved the error handling and model compatibility and updated the CTranslate2 version.
deep-learninginferenceopenaiquantizationspeech-recognition
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.