At JabRef, I built an OCR feature that replaced a manual external-tool and relinking workflow with a single “Perform OCR” menu action.
I designed an extensible OcrEngine interface and shipped integrations for OCRmyPDF and Docling. I also built a text-embedding pipeline that writes searchable text back into PDFs.
For my Document Search Engine project, I built a Java search engine from scratch, indexing a 100K-document corpus with a custom inverted index and TF-IDF ranking.
I also developed a Flask API for football match results using scraped data. In the JabRef community, I mentored new contributors and reviewed pull requests.

