Nic Weber and Kyle Niemeyer
April 25, 2025
We’re thrilled to announce the first two recipients of the US Research Software Sustainability Institute (URSSI) Early-Career Fellowship. This program supports emerging researchers to advance scientific software development practices.
Matthew Feickert will develop an open-source course on creating reproducible software environments for scientific and AI/ML applications. His project “Reproducible Machine Learning Workflows for Scientists” will focus on using tools like Pixi to create version-controlled research environments with specialized hardware accelerator support. Matthew will organize a workshop for researchers across scientific domains and also host a tutorial on Pixi at the 2025 SciPy conference.
Sam Zhang will address scientific reproducibility through his project “Principled Data Processing for Scientific Reproducibility.” Based on Principled Data Processing methodology, Sam plans to build a Python package that improves data analytic workflows by solving key challenges in project initialization, cross-platform collaboration, relative path management, data version control, and integration with supercomputing clusters. His goal is to develop a tool that becomes as commonplace for researchers as git, targeting 100 active users in the 2025-2026 academic year.
We’re excited to have Matthew and Sam working with us over the next few months, and look forward to the impact they will each have on two unique research software domains.
Watch for our next call for fellowship proposals this Spring (2025). We’ll continue seeking projects focused on AI/ML integration in scientific software, software sustainability best practices, and software education research.