Sebastian Wrede is a software engineer and PhD researcher specializing in privacy-preserving machine learning, with a particular focus on federated learning. He has eight years of experience bridging research and production, implementing a federated backend for Apache SystemDS and contributing backend features like feature hashing and aggregation rewrites to that widely used open-source ML system. Based in Copenhagen, he combines academic rigor from TU Graz and the IT University of Copenhagen with hands-on engineering at Sopra Steria. His work spans distributed ML, data management, and scalable systems, and he has practical experience deploying models where data cannot be centralized. Beyond code, he mentors students and has taught large-scale data analysis and big data management, reflecting a talent for translating complex concepts into practical exercises. Colleagues value him for turning privacy-focused research into maintainable, production-ready implementations.
8 years of coding experience
4 years of employment as a software developer
Doctor of Technical Sciences, Computer Science, Doctor of Technical Sciences, Computer Science at Graz University of Technology
Primary and Lower Secondary School, Primary and Lower Secondary School at Strandskolen
The Higher Commercial Examination (HHX), Elite-Course, The Higher Commercial Examination (HHX), Elite-Course at Niels Brock
Bachelor of Science, Software development, Bachelor of Science, Software development at IT University of Copenhagen
An open source ML system for the end-to-end data science lifecycle
Role in this project:
Backend Developer
Contributions:75 reviews, 62 commits, 77 PRs in 2 years 11 months
Contributions summary:Sebastian's commits primarily focused on extending the transform encode/apply functionality within the Apache SystemDS repository. These changes involved feature hashing, specifically within the `EncoderMVImpute` and `MultiReturnParameterizedBuiltinSPInstruction` classes. Furthermore, the user was responsible for adding rewrites related to aggregations, specifically focusing on the removal of unnecessary RemoveEmpty and CTable operations. This work involved modifying various Java files and test scripts.
An open source ML system for the end-to-end data science lifecycle
Contributions:4 PRs, 37 pushes, 4 branches in 4 months
pythonend-to-endml-systemlifecycledata-science
Find and Hire Top DevelopersWeโve 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.