Scholar
Thomas Wolf
Google Scholar ID: D2H5EFEAAAAJ
Co-founder at HuggingFace
Citations & Impact
Citations
46,390
H-index
38
i10-index
56
Publications
20
Co-authors
45
list available
Publications
Resume (English only)
Academic Achievements
- Led the creation of the Hugging Face Transformers and Datasets open-source libraries
- Co-authored the O'Reilly book 'Natural Language Processing with Transformers'
- Authored 'The Ultra-scale Playbook' on large-scale AI training
- Maintains a technical blog with posts accumulating over 250,000 views by end of 2018
- Produces educational videos such as 'The Future of Natural Language Processing'
Research Experience
- Conducted research on laser-plasma interactions at the BELLA Center, Lawrence Berkeley National Laboratory
- Worked as a Patent Attorney at Cabinet Plasseraud for five years, advising startups and large corporations on intellectual property
- Began consulting for Deep Learning/AI/ML startups in 2015, reigniting his interest in ML/DL
- Co-founded Hugging Face and led the development of the Transformers and Datasets libraries
- Co-organized the BigScience Workshop on Large Language Models, leading to the BLOOM model and dataset
Background
- Co-founder and Chief Science Officer (CSO) of Hugging Face
- Instrumental in initiating Hugging Face’s open-source, educational, and moonshot efforts
- Passionate about building open-source software to make complex research, models, and datasets widely accessible
- Advocates for open science in AI/ML and bridges the gap between academia and industrial labs
- Current research interests focus on the future of AI and moonshot initiatives
- Enjoys creating educational content on AI, ML, and NLP
