Sebastian Pütz
Senior Software Engineer at ENLYZE GmbH
Karlsruhe, Baden-Württemberg, Germany
Summary
🤩
Rockstar🎓
Top SchoolSebastian Pütz is a Senior Software Engineer based in Karlsruhe with eight years of experience building production Rust applications across industrial automation, speech processing and FinTech. He combines an academic background in computational linguistics and NLP with hands-on systems work—shipping edge-device software and cloud services alike. At companies from ENLYZE to Mercedes-Benz and Dock Financial he has driven architecture, observability improvements and pragmatic integrations for card processing and digital wallet systems. An active open-source contributor, he has optimized core tokenization performance in the widely used HuggingFace tokenizers and hardened PyO3’s Rust↔Python conversions. He routinely bridges research and engineering, translating NLP insight into robust, high-performance implementations. Colleagues describe him as a pragmatic problem-solver who favors efficient algorithms and operational resiliency.
8 years of coding experience
6 years of employment as a software developer
Admission as Doctoral Student, Computational Linguistics, Admission as Doctoral Student, Computational Linguistics at University of Tübingen
German, English, Chinese
Stackoverflow
2,530reputation
100kreached
76answers
1question
rust
top-5%
Github Skills (21)
algorithm10
transformers10
algorithms10
tokenize10
python10
python-c-api10
data-structure10
binding10
tokenizer10
ffi10
natural-language-processing10
data-structures10
nlp10
rust10
bert9
Programming languages (7)
Github contributions (5)
PyO3/pyo3
Oct 2019 - Sep 2020
Rust bindings for the Python interpreter
Role in this project:
Back-end Developer
Back-end DeveloperContributions:23 reviews, 30 commits, 14 PRs in 10 months
Contributions summary:Sebastian focused on enhancing the PyO3 library, a Rust binding for the Python interpreter. They fixed derive implementations for protocols like `PyObjectSetAttrProtocol` and removed methods from `PyMappingProtocol` based on CPython behavior. They also added type information to conversion errors and refactored code related to conversions, and the internal structure of `PyCell`. The user's contributions primarily revolved around improving the library's functionality and correcting internal implementation details.
pythonpython-c-apirustpython-interpreterrust-bindings
huggingface/tokenizers
Jul 2020 - Oct 2020
💥 Fast State-of-the-Art Tokenizers optimized for Research and Production
Role in this project:Back-end Developer
Back-end DeveloperContributions:2 reviews, 10 PRs, 29 comments in 3 months
Contributions summary:Sebastian focused on optimizing the BPE (Byte Pair Encoding) tokenizer, improving performance through caching and efficient merging strategies. They also refined various normalization methods, making them more efficient by leveraging features like `char` copy semantics and double-ended iterators. Furthermore, the user addressed issues related to tokenization, including fixes for training panics and enhancements to iterators.
nlptransformer-architecturelanguage-modelnatural-language-understandingtransformers
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.