2026 ASEE Annual Conference & Exposition

This paper presents a scalable framework for implementing open-ended group projects in large-enrollment CS1 courses. With the increasing availability and capabilities of generative AI tools, it has become essential to rethink how we design curriculum and assessments for introductory programming students. Many traditional CS1 projects can now be easily completed using freely available AI tools, raising concerns about academic integrity and learning outcomes. Additionally, increasing the weight of traditional assessments may heighten anxiety and disproportionately disadvantage students who are less prepared for high-stakes testing. The approach described in this paper offers an alternative model for integrating core programming concepts through open-ended, collaborative projects. While managing such projects at scale poses logistical and assessment challenges, this work outlines detailed procedures for project design, team formation, grading, and student support. Projects are introduced to students in a hackathon-style format, with groups selecting from one of three thematic tracks. Teams are formed based on student ability levels and assigned a TA who serves as a project manager throughout the multi-week experience. Lab sessions in the final weeks of the semester are used for progress updates, with grading based on iterative development and team collaboration. In the final week, students present their projects, and selected showcase teams are allowed to substitute their project for a final exam grade. This model has been implemented over 10 semesters and has shown to be a motivating and rewarding way to conclude the course for both students and instructors. The paper includes student feedback, project examples, and practical insights for instructors looking to adopt or adapt this format.