Felix Sonntag
Lead Software Engineer at Agoda
Bangkok, Thailand
Summary
👤
Senior🎓
Top SchoolFelix Sonntag is a Lead Software Engineer with 11 years’ experience specializing in machine learning and natural language processing, currently driving ML initiatives at Agoda from Bangkok. He combines a strong academic foundation from TUM and international research stints with hands-on engineering across Python, Kotlin/Java and mobile iOS, enabling end-to-end delivery from data pipelines to production features. Felix has deep experience scaling data systems and search quality—having worked on knowledge-graph indexing and performance improvements that increased concurrency by an order of magnitude. He also contributes to open-source ML tooling, including fixes and tutorial updates for Snorkel’s weak supervision notebooks, reflecting a pragmatic focus on reproducibility and developer experience. Colleagues rely on him for turning research ideas into robust, user-facing solutions while mentoring teams through technical complexity.
11 years of coding experience
9 years of employment as a software developer
Special Research Student Computational Linguistics, Special Research Student Computational Linguistics at Nara Institute of Science and TechnologyComputer Science, Computer Science at National Taiwan UniversityMaster of Science (M.Sc.) Informatics, Master of Science (M.Sc.) Informatics at Technical University of MunichComputational Linguistics, Computational Linguistics at Ludwig-Maximilians-Universität München
German, English, Chinese, Spanish, Japanese
Github Skills (12)
machine-learning10
jupyter-notebook10
python10
snorkel10
data-science10
basics9
learn-ruby-on-rails9
user-manual9
nlp8
scikit7
scikit-learn7
data-engineering6
Programming languages (15)
Github contributions (5)
snorkel-team/snorkel
May 2017 - Jun 2017
A system for quickly generating training data with weak supervision
Role in this project:
Data Scientist
Data ScientistContributions:7 commits, 1 PR, 14 comments in 21 days
Contributions summary:Felix's contributions center around modifying and extending Jupyter Notebooks related to a weak supervision system. Their changes include adapting existing code for compatibility and fixing unicode issues to ensure smooth execution. The user also appears to be working with tutorials, modifying code for existing components to update the code and correct small errors. Furthermore, they made changes to imports and file interactions.
weak-supervisionpythondata-sciencemachine-learninglabeling
fsonntag/snorkel
Apr 2017 - Jul 2018
Contributions:142 pushes, 4 branches, 3 tags in 1 year 3 months
management-systemextractionpythondata-sciencemachine-learning
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.