⚡ Website ✍ Blog ☕ Coffee ❤️ Sponsor
Senior Research Data Engineer based in Berlin, specializing in NLP and AI.
15+ years of building NLP pipelines and large-scale datasets, from German web corpora for research to multilingual data for LLM and Voice AI training. PhD in computational linguistics.
I like to build in the open. My packages include Trafilatura, a web text extraction library with 7K+ stars and 13M monthly downloads, described in an ACL 2021 paper.
It is used across thousands of projects, including by Hugging Face, IBM, Microsoft Research, NVIDIA, the Allen Institute for AI, Stanford and the Internet Archive.
- The Art of Extraction: Trafilatura's Connection to Pasta
- Respectful Web Crawling and Downloads with Courlan
- Multilingual Topic Modeling with Simplemma
- Extracting the main text content from web pages using Python
- A simple multilingual lemmatizer for Python
- A module to extract date information from web pages