GeomScale Project

GeomScale is a NumFOCUS-affiliated research and development project delivering open-source software for state-of-the-art algorithms at the intersection of data science, optimization, geometric computing, and statistical computing. We are a Google Summer of Code mentoring organization and part of the global scientific Python and open-source ecosystems.

Our focus is on scalable algorithms for sampling from high-dimensional distributions, volume approximation, convex optimization, integration, and related tasks that are foundational for modern data science and machine learning. We bridge the gap between cutting-edge theory in high-dimensional geometry and optimization and robust, production-quality implementations that researchers and practitioners can rely on.

Applications of GeomScale span computational biology (metabolic networks and systems biology), finance (portfolio optimization, risk and shock detection in large markets), Bayesian statistics, AI, and applied mathematics.