Tom De Geus is a patent specialist and former senior researcher in complex systems physics with 11 years of experience combining deep computational research and robust software engineering. He led independent, funded research at EPFL (Ambizione fellowship) and built high-performance simulation and data pipelines, supervising students and contributing to top journals and conferences. Equally at home in C++ and Python, he is an active open-source contributor to impactful projects such as HighFive (C++ HDF5 interface), xtensor, and doxygen, adding features, tests, and documentation that improve scientific reproducibility and tooling. His work bridges academia and industry—translating research-grade code into production-ready tools and even commercial valorization projects for industrial partners. Now based in Lausanne, he applies this technical depth to intellectual property work, offering a rare mix of hands-on code expertise, scientific rigor, and patent analysis. An unexpected strength is his track record of turning numerical methods and data-heavy research into clear, maintainable libraries that others can adopt.
11 years of coding experience
5 years of employment as a software developer
Doctor of Philosophy (Ph.D.), Mechanical Engineering, Cum Laude, Doctor of Philosophy (Ph.D.), Mechanical Engineering, Cum Laude at Eindhoven University of Technology
Post-master, Egineering Mechanics, Post-master, Egineering Mechanics at Graduate School Engineering Mechanics
Contributions:3 releases, 79 reviews, 29 commits in 3 years 1 month
Contributions summary:Tom focused on developing and expanding the HighFive library, a header-only C++ interface for HDF5. Their contributions involved implementing a high-level API (H5Easy) for data import, export, and format conversion, including support for various data types like scalars, vectors, xtensor arrays, and Eigen matrices. They added support for n-dimensional vectors, OpenCV and expanded the API to include compression options and attributes. The user also simplified existing xtensor and Eigen implementations.
Contributions:147 reviews, 106 commits, 177 PRs in 4 years 6 months
Contributions summary:Tom primarily contributed to the documentation and core functionality of the C++ tensor library. They added missing headers, fixed compiler warnings, and implemented the "interp" function, which is a one-dimensional piecewise linear interpolant. The user also implemented functionality for set difference operations and added test cases for those operations. In addition to functionality, the user updated existing documentation and added further NumPy comparisons.
cppmpinumpypython-bindingsc-plus-plus
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.