Klymkowsky will speak about the innovative ways he and his collaborators utilize AI Bots in their classes. His Bots, which he named Dewey and Rita, act like “tutors” for his students. They also can analyze students’ work and instructors’ questions, providing feedback and suggestions to instructors and course designers that can help students overcome conceptual obstacles and help enhance their learning skills and outcomes. Recent surveys show that a large majority of both undergraduate and graduate students are using Artificial Intelligence tools, encouraging the development of new ideas on how to incorporate generative AI strategically into college and university coursework.
Scientists, Engineers, and Technologists live in a world of data, methodical analysis, and measurable outcomes. Scientific research is increasingly being held to regulatory standards that are vague and ambiguous. How can the scientific community understand and adapt to this shifting regulatory landscape?
This panel will explore diverse approaches to integrating AI in teaching and research. Each panelist will briefly share their current work with AI before opening up to an informal, generative discussion with the audience about the opportunities and challenges AI presents in higher education. Lee Frankel-Goldwater (Environmental Studies) will discuss the AI Literacy Ambassadors program and the use of citizen science lessons to build research and AI literacy skills. Bobby Hodgkinson (Aerospace Engineering) will share insights from embedding AI-generated feedback in engineering courses, highlighting both its pedagogical strengths and limitations. Diane Sieber (Engineering, Ethics and Society) will focus on shifting campus culture through new curricula and the Generative Futures Lab for AI — a collaborative space for experimentation and knowledge-sharing. Together, their work spans interdisciplinary areas of application, classroom integration, and institutional transformation — offering a rich conversation on the evolving role of AI in academia.
Microsoft Windows through Open OnDemand by Paige Despain 7lbd is an innovative open-source project that simplifies Windows deployment in HPC environments by treating Windows as an application within Open OnDemand. The solution eliminates traditional infrastructure complexities by using technologies like Apache Guacamole, network namespaces, and a simplified Windows VM configuration to provide secure, isolated Windows desktops across computing clusters. This solution simplifies Windows to a level that even Linux systems administrators will find easy to maintain while maintaining robust security and accessibility, no AD required.
An Approach to SLURM Configuration Verification by Kyle Reinholt Ensuring the correctness of Slurm configurations is crucial for maintaining high-performance computing (HPC) environments, but validating these configurations effectively remains a challenge. In this lightning talk, we will explore existing approaches to Slurm configuration verification, including manual checks, custom scripts, and automated validation tools. While these methods offer some benefits, they often fall short in scalability and flexibility. The talk will then shift focus to exploring potential solutions for improving configuration verification, discussing innovative strategies, and tools that could streamline the process, reduce errors, and enhance cluster reliability.
