All workTeam Lead · Full Stack Developer · Graduation Project

Gallifreyan Page

Sustainable bookstore built around second-hand books and recycling, aligned with the UN SDGs. Graduation project.

20241 links
gallifreyan-page
Gallifreyan Page

Gallifreyan Page is a full-stack e-commerce platform for books and music CDs, built by a team of three as our software engineering graduation project; I led the team and built it full stack. It pairs a second-hand section and a book and paper recycling program that pays users back in store balance with educational content on sustainability, plus a GPT-4o writing assistant for authors. The frontend is Angular with RxJS; the backend is .NET following Clean Architecture, with MongoDB, RabbitMQ for asynchronous work and JWT authentication.

Context

A graduation project with a sustainability brief. Books are one of the easiest things to reuse, yet new copies keep getting more expensive and old ones end up in the bin. We designed the project around the United Nations Sustainable Development Goals: Goal 12 (Responsible Consumption and Production), Goal 4 (Quality Education) and Goal 13 (Climate Action).

Problem

Make books cheaper to get and keep more of them in circulation, without turning sustainability into a slogan on the homepage. The store had to give people a real reason to bring back the books they no longer read, and a cheaper way to buy the ones they want.

Approach

We were a team of three. I led the team, designed the screens in Figma and built both the frontend and the backend; one teammate owned testing, and the other handled data analysis with Amplitude. The store sells new releases next to a dedicated second-hand section. Users can sell or donate old books and paper through the recycling program; what they bring back turns into store balance they can spend on their next book. A sustainability section explains the SDGs and how each feature maps to them, and the site features Twin Science and Carbon Gate to point readers toward STEM education and carbon-neutral initiatives. For writers, a GPT-4o assistant suggests story continuations and helps with writer's block. On the backend, a Clean Architecture .NET API separates domain, application and infrastructure layers, and RabbitMQ takes slow work off the request path.

Outcome

Delivered in 2024 as our graduation project for the B.S. in Software Engineering at Istanbul Aydın University, covering the full flow from browsing and the basket to orders, the recycling form and the AI assistant. The source code is public on GitHub.

© 2026 Berkay Arda Yıldız