Maarten Van Gompel
Research Software Engineer at Radboud University Nijmegen
Eindhoven, North Brabant, Netherlands
Summary
👤
Senior🎓
Top SchoolMaarten Van Gompel is a Research Software Engineer with 16+ years specializing in language technology, NLP and open-source tooling, currently working across CLARIN/CLARIAH projects at KNAW and Radboud University. He combines deep academic research (PhD on constructions for machine translation) with hands-on engineering in Python, C/C++, Rust and shell, contributing to production-grade tools like Frog, Moses and tokenization/NER pipelines. Maarten is an active open-source maintainer and contributor—his work on token classification for the popular rust-bert project and contributions to foundational text-processing libraries demonstrate both model-level and systems-level expertise. He prioritizes Linux, reproducibility and InfoSec practices, and prefers to engage via his personal site, email and federated social accounts rather than LinkedIn. Colleagues value him for turning linguistic research into reliable, reusable software components and for bridging research and production deployment. An understated but telling detail: he runs and curates a personal research/software hub that centralizes his code, papers and long-form technical writing.
16 years of coding experience
2 years of employment as a software developer
Bachelor, Cognitieve Kunstmatige Intelligentie (CKI) / Cognitive Artificial Intelligence, Bachelor, Cognitieve Kunstmatige Intelligentie (CKI) / Cognitive Artificial Intelligence at Utrecht UniversityMaster, Human Aspects of Information Technology / Computational Linguistics, Master, Human Aspects of Information Technology / Computational Linguistics at Tilburg UniversityHigh-school, Natuur en Gezondheid, Natuur en Techniek, Maths, Physics, Chemistry, Biology, High-school, Natuur en Gezondheid, Natuur en Techniek, Maths, Physics, Chemistry, Biology at Heerbeeck College, Best
Dutch, English, Spanish, Portuguese, French, German, Italian, Esperanto, Russian, Arabic, Chinese
Stackoverflow
505reputation
43kreached
28answers
1question
Github Skills (32)
debugging10
tokenize10
machine-translation10
debug10
c-language10
python10
machine-learning10
data-structure10
tokenizer10
natural-language-processing10
bert10
data-structures10
transformer10
nlp10
rust10
Programming languages (26)
Github contributions (5)
fbkarsdorp/python-course
Mar 2013 - Apr 2013
Tutorial and introduction into programming with Python for the humanities and social sciences
Role in this project:
Back-end Developer
Back-end DeveloperContributions:130 commits in 26 days
Contributions summary:Maarten primarily contributed to the development of a text analysis application using Python. Their work involved the implementation of core components, including tokenization, sentence splitting, and n-gram extraction, as well as the creation of a frequency list. The user also integrated these components and incorporated command-line argument parsing, culminating in a functional text processing tool. These changes built the foundational preprocessing steps of the Python course.
pythonsocial-scienceshumanitiessciencespython3
moses-smt/mosesdecoder
Aug 2012 - Jun 2014
Moses, the machine translation system
Role in this project:Back-end Developer
Back-end DeveloperContributions:18 commits in 1 year 9 months
Contributions summary:Maarten primarily debugged and improved the Moses machine translation system. Their contributions focused on refining the `Hypothesis` class, which is central to the decoding process, likely involving modifications to scoring and feature calculations. The commits indicate attempts to identify and correct issues related to score breakdowns and feature interactions. This involved changes in the source code, specifically within the `Hypothesis.cpp` and `Hypothesis.h` files, improving the quality of machine translation results.
i18nmachine-translationtranslationlocalizationmoses
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.