Ella Charlaix
Machine Learning Engineer at Hugging Face
Paris, Ile-de-France
Summary
🤩
Rockstar🎓
Top SchoolElla Charlaix is a Machine Learning Engineer based in Paris with five years of experience building and optimizing production-ready NLP models, currently contributing at Hugging Face. She specializes in model export, inference acceleration and compression techniques—ONNX export, quantization, pruning and distillation—and has driven package architecture and tooling for Optimum-Intel and Optimum to streamline deployment on CPU backends. Her contributions to the flagship transformers library include enabling ONNX workflows for multiple-choice tasks and improving compatibility across runtimes, reflecting a strong focus on robustness and interoperability. Trained in robotics and computational science at EPFL, she brings a systems-oriented, research-backed approach to model optimization and a track record of turning experimental methods into practical deployment scripts and notebooks. An understated strength is her attention to developer ergonomics—clearing warnings, improving examples, and fixing CI/docker issues to make advanced ML tooling more accessible.
5 years of coding experience
3 years of employment as a software developer
Bachelor of Science (B.Sc.), Mechanical Engineering, Bachelor of Science (B.Sc.), Mechanical Engineering at Ecole polytechnique fédérale de LausanneBaccalauréat Scientifique, Spécialité Mathématiques, Mention Européenne (anglais), Baccalauréat Scientifique, Spécialité Mathématiques, Mention Européenne (anglais) at Cité Scolaire Internationale de Lyon
French, English, German
Github Skills (24)
transformers10
pytorch10
quants10
intel10
python10
machine-learning10
distill10
inference10
onnx10
pruning10
onnxruntime10
model-optimization10
quantization10
optimization10
openvino9
Programming languages (4)
Github contributions (5)
huggingface/optimum-intel
May 2022 - Jan 2023
🤗 Optimum Intel: Accelerate inference with Intel optimization tools
Role in this project:
ML Engineer
ML EngineerContributions:41 releases, 1292 reviews, 129 commits in 8 months
Contributions summary:Ella's commits focus on setting up and refactoring the `optimum-intel` package, which is designed to accelerate inference using Intel optimization tools. They established the package structure by setting up the necessary files such as `setup.py` and version files. Furthermore, the user implemented example scripts related to language modeling and question answering. The user also introduced and integrated distillation, pruning, and quantization techniques, indicative of a strong focus on model optimization and deployment.
pytorchtransformersdeep-learningpruningoptimization
huggingface/optimum
Jul 2021 - Jan 2023
🚀 Accelerate inference and training of 🤗 Transformers, Diffusers, TIMM and Sentence Transformers with easy to use hardware optimization tools
Role in this project:Back-end & ML Engineer
Back-end & ML EngineerContributions:29 releases, 549 reviews, 125 commits in 1 year 6 months
Contributions summary:Ella's primary contribution focused on adding features for exporting transformer models to ONNX, optimizing them with ONNX Runtime, and integrating dynamic quantization. This included creating scripts for model conversion and optimization using onnxruntime.transformers, and adding various options for controlling the optimization process. They also introduced dynamic quantization for optimizing and reducing the size of the model.
trainingoptimization-toolstransformersintelhardware
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.