Patrick Varilly is a freelance software engineer with 14 years of experience building scalable, production-grade systems for research labs and Fortune 500 customers from Brussels. He specializes in data engineering, backend development and performance-critical computation, with a recent focus on scaling Bayesian phylogenetic inference and production-hardening tools for genomic epidemiology at the Sabeti Lab. Patrick has a strong track record shipping robust distributed systems and DSLs for large-scale data migration and search (e.g., DobiMigrate and a trademark search serving ~200M records), and contributes to scientific open-source projects including notable work improving SciPy’s spatial cKDTree accuracy and tests. He blends academic rigor (postdoc and PhD-level research in computational physics/chemistry) with pragmatic engineering practices like TDD, immutable data structures and asynchronous libraries. Known for surfacing high-impact analyses and turning R&D prototypes into reliable services, he moves fluidly between C++, Java, Python and TypeScript to remove scaling bottlenecks. An understated strength is his ability to visualise and explain complex computational engines, making inner workings accessible to both researchers and engineers.
14 years of coding experience
10 years of employment as a software developer
University of California, Berkeley
Bachelor of Science (BS), Physics (minor in Math), 5.0 / 5.0, Bachelor of Science (BS), Physics (minor in Math), 5.0 / 5.0 at Massachusetts Institute of Technology
Back-end Developer & QA Engineer / Test Automation Engineer
Contributions:26 commits, 3 comments in 1 year 4 months
Contributions summary:Patrick primarily contributed to the `scipy/scipy` repository by addressing bugs and implementing improvements in the `scipy.spatial.kdtree` and `scipy.spatial.ckdtree` modules. Their work included bug fixes for the `kdtree`, and adding and enhancing functions such as `query_ball_point`, `query_ball_tree`, `query_pairs`, and `sparse_distance_matrix` to the `ckdtree` module. The user also focused on ensuring code quality by addressing issues related to accuracy and providing testing infrastructure enhancements to the cKDTree implementation.
Python module for calculating DNACC interaction potentials and binding details
Contributions:9 commits, 3 PRs, 4 pushes in 8 years 4 months
interactionpythonpython-modulepotentialsdetails
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.