Implementation of Clean Architecture in the Back-end of an Alumni Program Application: A Case Study of a Non-profit Organization

Abstract

Alumni data tracking within the non-profit organization located in South Jakarta previously relied on Google Forms, which presented significant limitations in dynamic data management and carried a high risk of data redundancy. Furthermore, an unstructured back-end codebase in the legacy internal system hindered long-term maintenance and system evolution. This study aims to build a highly structured back-end system using the Clean Architecture pattern for an alumni program application to facilitate efficient, scalable, and testable data management. The system was developed using Node.js and the Hapi.js framework to deliver a robust RESTful API. The development process was guided by the Scrum methodology across eight sprints, where Sprints 1–4 focused on database design using the Database Life Cycle (DBLC) method through normalization up to the third normal form (3NF), and Sprints 5–8 focused on back-end component development. Clean Architecture separates system responsibilities into four primary horizontal layers: Domain, Applications, Interfaces, and Infrastructures. The results indicate that the designed MySQL relational database successfully addresses university and domicile data redundancy through a centralized schema. Functional validation via unit, integration, and system testing demonstrated that all functional requirement scenarios were fully satisfied. Architectural validation using an Architecture Fitness Function based on stability and abstractness metrics proved full compliance with The Dependency Rule, where the Domain layer was empirically shown to be the most stable ($I=0.04, A=0.62$) and the Infrastructures layer was concrete ($I=0.92, A=0.00$). Consequently, the implementation of Clean Architecture successfully establishes a robust, modular, and future-ready back-end foundation.