Interdisciplinary Polymer and Data Science Graduate Training by Bringing Contemporary Industry Problems into the Classroom

The increasing demand for sustainable, functionally optimized polymers requires a fundamental shift away from traditional trial-and-error towards AI- and data-driven approaches. This shift creates a critical need for an R&D workforce equipped with foundational expertise in polymer science, complemented by interdisciplinary training in computing and data science. To address this need, we have developed the NSF NRT-MIDAS "Hackathon" course at the University of Delaware. This semester-long graduate-level course is designed as an intensive experiential learning opportunity, challenging students to tackle real-world polymer and soft materials problems from industry and national laboratory partners (e.g., Dow, DuPont, Merck). Students from diverse backgrounds, including Chemical Engineering, Materials Science Physics, Chemistry, and Computer/Data Science, are intentionally organized into interdisciplinary teams of 3–4 members. These teams work on multifaceted challenges that span machine learning, molecular modeling, and data analysis. Through regular interactions with industry professionals, students gain an appreciation for applying computational tools to solve practical materials challenges. The collaborative, problem-driven nature of the course fosters essential non-technical skills, including adaptive communication, leadership, and team management. This course serves as a replicable model integrating Data Science into Polymer Science education to meet the demands of modern, accelerated materials innovation.