Severyn Kozak is a systems programmer and quantitative developer with 13 years of experience building high-performance trading and data systems, grounded in MIT computer science and math training. He has delivered low-latency software for equities, futures, and crypto at firms like Ansatz Capital and Tower Research, pairing C++ core infra work with quant-driven engineering. His open-source contributions to projects such as Mapzen/Pelias and Tilezen demonstrate practical expertise in geospatial databases and PostGIS performance tuning, including writing SQL functions and indexes to scale vector tile pipelines. Severyn combines deep performance optimization skills with a strong background in automated testing and security tooling from internships at MongoDB and Cryptosense. Based in New York, he blends rigorous academic foundations with hands-on production experience across trading systems and geospatial data stacks. An understated strength is his ability to translate mathematical rigor into pragmatic, measurable improvements in throughput and query performance.
13 years of coding experience
4 years of employment as a software developer
Computer Science (course 6-3), minor in Math (course 18), Computer Science (course 6-3), minor in Math (course 18) at Massachusetts Institute of Technology
Diploma, General, Diploma, General at Stuyvesant High School
Contributions:1 release, 51 commits, 13 PRs in 19 days
Contributions summary:Severyn primarily contributed to the back-end of the Pelias API, focusing on enhancing search result accuracy and performance. They implemented exact text match boosting, improved the sorting of results, and fixed related test failures. Furthermore, they addressed code quality issues, refactoring components, and improving error handling for internal server errors, indicating a focus on code maintainability and robustness.
Tilezen vector tile service - OpenStreetMap data in several formats
Role in this project:
Database Engineer / Database Administrator
Contributions:23 commits, 6 PRs, 21 pushes in 9 months
Contributions summary:Severyn primarily focused on enhancing the `tilezen/vector-datasource` repository with new database functions and related improvements. The commits involve creating and modifying SQL functions for splitting polygons into tiles, optimizing data handling for the Natural Earth dataset, and improving query performance. They also created indexes and sequences for efficient data management within the PostGIS database.
pythonfltkmap-editortilesetmapzen
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.