Group work in high school classes often faces challenges like unequal participation and poor team dynamics, but these issues are particularly significant in coding classes. Collaboration is a core component of CS and CSed, where students must work together to solve problems, debug, and manage projects. To address these challenges, this talk introduces a conceptual AI-driven tool, CollabCode, specifically designed to monitor and enhance group work in high school coding classes. Collabcode uses machine learning algorithms to track individual student participation, task distribution, and communication patterns in real-time. Based on this data, the system provides personalized feedback to students and generates actionable insights for teachers. The tool can suggest appropriate roles or task assignments based on real-time data, helping students demonstrate and enhance their skills in different capacities. By identifying patterns of teamwork such as disengagement or dominance, Collabcode can recommend equitable group structures. Teachers receive detailed collaboration analytics that suggest how tasks can be distributed to maximize each student’s contribution and foster more balanced cooperation. The COVID-19 pandemic exacerbated the issues of group work by reducing opportunities for hands-on teamwork and socialization. This tool aims to help rebuild those skills. Through dynamic task and role management, CollabCode ensures that group work is more productive, allowing students to develop both their technical and collaboration skills, which are critical in and outside of computer science classrooms.