Felix Dittrich
Lead ML Engineer at t2k GmbH
Greater Leipzig Area Germany
Summary
🤩
Rockstar🎓
Top SchoolFelix Dittrich is a Lead ML Engineer with eight years of experience building production-ready machine learning systems and establishing ML capabilities from scratch within companies. He maintains and contributes to open-source OCR tooling—most notably as a maintainer on mindee/doctr and his own OnnxTR project—demonstrating hands-on expertise in document text recognition and model deployment. At t2k GmbH he progressed from ML Engineer to Lead ML Engineer, applying practical skills across Python-based deep learning, NLP, and computer vision while ramping up C++ as a second core language. Felix blends applied research curiosity with MLOps pragmatism, shipping features like hOCR export and robust dimension handling in upstream libraries. Based in the Greater Leipzig Area, he pairs academic grounding in computer science with a track record of improving developer-facing ML tooling.
8 years of coding experience
2 years of employment as a software developer
Bachelor of Science - BS, Computer Science, Bachelor of Science - BS, Computer Science at HTWK Leipzig - Hochschule für Technik, Wirtschaft und Kultur Leipzig
Stackoverflow
5reputation
1kreached
0answers
3questions
Github Skills (9)
ocr10
pytorch10
deep-learning10
ocra10
python10
xml9
tensorflow7
opencv6
fastapi6
Programming languages (9)
Github contributions (5)
mindee/doctr
Nov 2021 - Oct 2022
docTR (Document Text Recognition) - a seamless, high-performing & accessible library for OCR-related tasks powered by Deep Learning.
Role in this project:
Backend Developer & Data Scientist
Backend Developer & Data ScientistContributions:5 releases, 677 reviews, 133 commits in 11 months
Contributions summary:Felix implemented an XML export method for the `DocumentBuilder`, adding the functionality to export document pages as XML in the hOCR format. They also addressed a bug related to dimension handling during the export process and made improvements in the documentation, including fixes to docstrings. The contributions demonstrate the user's involvement in building and enhancing document processing functionalities within the doctr library, specifically focusing on OCR-related tasks.
ocr-recognitiontext-documentoptical-character-recognitiontext-recognitioncomputer-vision
felixdittrich92/OnnxTR
May 2024 - Mar 2025
OnnxTR a docTR (Document Text Recognition) library Onnx pipeline wrapper - for seamless, high-performing & accessible OCR
Contributions:14 releases, 2 reviews, 74 PRs in 10 months
deep-learningdocument-recognitionocronnxonnxruntime
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.