Builders & open source
Leandro von Werra
Machine learning engineer in the open-source team, Hugging Face
- Machine learning engineer in the open-source team, Hugging Face12
Leandro von Werra works in machine learning at Hugging Face, according to von Werra's GitHub profile, and Hugging Face's LLM Course describes von Werra as a machine learning engineer in the company's open-source team. Von Werra is listed first among the authors of TRL, Hugging Face's Apache-2.0 library for post-training transformer models, and is a co-author of the StarCoder and StarCoder2 code model papers. Von Werra also appears in the citation for smolagents, Hugging Face's library for running agents in a few lines of code, and co-wrote the O'Reilly book Natural Language Processing with Transformers.123456
At a glance
- Current role
- Machine learning engineer in the open-source team
- Company
- Hugging Face
- Sector
- Company
- Work listed
- 7
- Sources cited
- 12
- Profile reviewed
Open-source work
Leandro von Werra is the first-listed author of TRL (Transformers Reinforcement Learning), whose citation file dates the project to March 2020. TRL is an Apache-2.0 library for post-training transformer models with methods such as supervised fine-tuning, reward modeling, Direct Preference Optimization and GRPO.3
Von Werra is also listed in the citation for smolagents, a Hugging Face library for running agents in a few lines of code. Hugging Face's LLM Course describes von Werra as a machine learning engineer in the open-source team with industry experience bringing NLP projects into production, and as a co-author of the O'Reilly book Natural Language Processing with Transformers.62
Code models and datasets
Von Werra co-authored a series of papers on open code models and their data. The Stack (2022) gathered permissively licensed source code in 30 programming languages with a data governance plan, and was followed by the SantaCoder (2023), StarCoder (2023) and StarCoder2 (2024) model papers.7845
Von Werra is also a co-author of FineWeb (2024), a pretraining dataset derived from Common Crawl snapshots, and of SmolLM2 (2025), which documents the data-centric training of a 1.7-billion-parameter small language model.910
Work and contributions
TRL (Transformers Reinforcement Learning) Open-source code
Listed first among the authors in TRL's citation file, which dates the project to 2020. TRL is an Apache-2.0 toolkit for post-training transformer models with SFT, reward modeling, DPO, and GRPO.3
StarCoder2 Published research
Co-author of "StarCoder 2 and The Stack v2: The Next Generation" (2024), and earlier of "StarCoder: may the source be with you!" (2023).54
smolagents Open-source code
Listed among the authors in the citation entry of the smolagents README (2025); the library runs agents in a few lines of code and is licensed under Apache 2.0.6
Natural Language Processing with Transformers (O'Reilly) Published research
Co-author of the O'Reilly book, according to the author biographies on Hugging Face's LLM Course page.2
BigCodeBench Open data
Last author of the BigCodeBench paper, a benchmark for code generation with diverse function calls and complex instructions that the repository README cites.11
FineWeb Open data
Co-author of the FineWeb datasets paper, which the FineWeb dataset card cites.9
StarCoder2-3B Open weights
One of the two corresponding authors, marked with a star, of the StarCoder 2 paper that introduced StarCoder2-3B.12
Timeline
- Released TRL, according to the project's citation file.3
- Co-authored The Stack, a dataset of permissively licensed source code.7
- Co-authored the SantaCoder code model paper.8
- Co-authored "StarCoder - may the source be with you!"4
- Co-authored the StarCoder2 and The Stack v2 paper.5
- Co-authored the FineWeb datasets paper.9
- Co-authored the SmolLM2 paper on training a small language model.10
Where they work
- Hugging Face 34 people in the wiki
Related people
- Younes BelkadaAlso at Hugging Face
- Edward BeechingAlso at Hugging Face
- Lewis TunstallAlso at Hugging Face
- Stas BekmanAlso at Hugging Face
- Philipp SchmidAlso at Hugging Face
- Noam BrownReinforcement learning · Language models
- Weizhu ChenLanguage models · Reinforcement learning
- Eric MitchellLanguage models · Reinforcement learning
- Nathan LambertLanguage models · Reinforcement learning
- Shengjia ZhaoLanguage models · Reinforcement learning
Suggest a correction or removal
Profiles cover public professional work only, and each statement cites a source. If something is out of date, missing or wrong, tell us and link a source. People can also ask for their profile to be corrected or removed.
References
12 sources · profile reviewed Oct 21, 2026 · data updated Oct 22, 2026
- lvwerra (Leandro von Werra) on GitHub
- Hugging Face LLM Course, Chapter 1 introduction (About the authors)
- huggingface/trl CITATION.cff (authors, license Apache-2.0)
- StarCoder: may the source be with you! (arXiv 2305.06161)
- StarCoder 2 and The Stack v2: The Next Generation (arXiv 2402.19173)
- huggingface/smolagents README (citation) and LICENSE (Apache-2.0)
- The Stack: 3 TB of permissively licensed source code (arXiv 2211.15533)
- SantaCoder: don't reach for the stars! (arXiv 2301.03988)
- The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale (arXiv 2406.17557)
- SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model (arXiv 2502.02737)
- BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions (arXiv 2406.15877)
- StarCoder 2 and The Stack v2 (arXiv HTML, author list)