Grapevine, Enhancing Course Selection Through Machine Learning: A Technical Overview of Grapevine’s AI-Driven System; The Impact of Algorithmic Thinking in Early Computer Science Education on Cognitive Development and Problem-Solving in Underrepresented Communities

My STS research and technical project are connected through a shared theme of how technological systems can be designed and implemented in ways that address educational inequity. Both projects revolve around exploring how education can be made more inclusive and equitable for students. While my technical project focuses on building a web application that uses AI to help students navigate course selection, my STS research explores the ethical obligations of educators and institutions to implement computer science education that is relational, emotionally responsive, and just.
In my technical project, I worked on developing Grapevine, a course recommendation web application that helps students find UVA classes tailored to their interests and academic needs. Built using retrieval-augmented generation (RAG) and natural language processing, Grapevine allows students to ask free-form questions about course prerequisites or requirements and receive personalized suggestions. My team and I designed a full stack solution: we collected data from various platforms such as Lou’s List and UREG, cleaned and processed it into a centralized database, and then built a large-language model (LLM) that ranks and returns the results based on user queries. The app also features a user-friendly frontend that helps students navigate through suggestions efficiently. The goal of this project was to help students, especially those unfamiliar with UVA’s advising structure, feel more confident in their course planning.
My STS research investigated the ethical and cognitive implications of introducing algorithmic thinking in computer science education for underrepresented students. Using Carol Gilligan’s Care Ethics as a framework, I analyzed how teacher-student relationships, emotional engagement, and institutional support systems shape the way marginalized students experience technical learning. Drawing from Joanna Goode’s case studies, I argued that educators have an ethical responsibility to provide culturally responsive instruction, validate students’ identities, and create emotionally supportive learning environments. My research highlighted how algorithmic thinking, when taught with care, becomes an act of empowerment for students. The critiques of care ethics, such as its scalability, are addressed and defended in my research by demonstrating how care-based teaching leads to measurable gains in enrollment and retention in Goode’s case studies.
Working on both these two projects together in the same semester enhanced the quality of both. My technical work on Grapevine helped me better understand how personalization and user experience play out in real-time, which added depth to my theoretical exploration of algorithmic thinking in my STS paper. On the other hand, the ethical framework I developed in my STS research made me more aware of the importance of designing technical tools that support equity and visibility. I became more intentional about designing Grapevine in a way that supports students from various backgrounds. Working on both projects in the same semester allowed me to approach this issue from both a theoretical and practical perspective, and ultimately made me more thoughtful in my technical designs.

School of Engineering and Applied Science

Bachelor of Science in Computer Science

Technical Advisor: Seongkook Heo

STS Advisor: Kathryn Webb-Destefano

Technical Team Members: William Giles, Tiger Zhang