Improving Capstone Team Outcomes through Dynamic Skill Matching and Preference Alignment¶
Authors: Brandon Pardi, Garret Castro, Michael Pisman, Avash Adhikari, Santosh Chandrasekhar
Venue: CSCI 2025 — 28th International Conference on Computer Science and Informatics (accepted)
Links: GitHub Repository
Abstract¶
Team-based projects are a cornerstone of engineering and computing courses, but unstructured team formation often leads to poor project outcomes due to misaligned student interests and inadequate skill coverage. This paper introduces a novel, three-stage methodology for creating effective student teams by integrating student preferences with project skill requirements. In the first stage, students complete a survey to report their project interests and self-assessed skills. Next, a Large Language Model (LLM) analyzes project descriptions to extract the necessary skills for each project's success. Finally, a dynamic assignment algorithm matches students to projects, simultaneously maximizing skill coverage and preference alignment. Preliminary evaluations show our approach produces teams with higher skill coverage and better preference satisfaction compared to random or manual assignment approaches.
Research Context¶
This work is a core output of the CSED Lab's OptiTeam project, which investigates how algorithmic and AI-driven approaches can improve team formation in project-based computing education. The three-stage methodology — preference elicitation, LLM-based skill extraction, and constrained optimization — sits at the intersection of the lab's interests in educational technology, human-centered design, and applied AI.
The paper targets a longstanding pain point in capstone courses: unstructured team formation that produces unbalanced teams and poor student outcomes. By replacing ad-hoc assignment with a transparent, data-driven pipeline, the CSED Lab aims to demonstrate measurable improvements in both team performance and student satisfaction.