Student dropout remains a persistent concern in Philippine higher education, with nearly four out of ten college students leaving their programs despite the implementation of free tuition policies. At Colegio de San Gabriel Arcangel (CDSGA), particularly within the College of Computer Studies (CCS), students continue to face academic, behavioral, and personal challenges that increase the risk of disengagement and eventual dropout. This capstone project presents the development of a Dropout Risk Predictor designed for second- year CCS students of CDSGA. The system utilizes a 40-item preemptive diagnostic examination that measures academic performance, behavioral patterns, and personal factors. The results are combined with students' grades and attendance records and processed using a weighted scoring model consisting of 40% academic, 35% behavioral, and 25% personal indicators, supported by a lightweight predictive algorithm. Based on this process, students are classified into four risk levels: No Risk, Low Risk, Moderate Risk, and High Risk. The system was developed as a user-friendly desktop application intended for use by guidance counselors and faculty members. It enables early identification of at-risk students, provides visual reports, allows profile printing, and supports timely intervention planning before academic difficulties become severe. System testing demonstrated that the predictor produced consistent and reliable outputs that aligned with actual faculty observations, indicating its effectiveness as a proactive student support tool.