Abstract: This study addresses the challenge of identifying students at risk of academic underperformance in a BS Information Technology program. Using a predictive analytics framework aligned with the Philippine Computer Society’s Information and Computing Accreditation Board (PICAB) Criterion 3 on Student Outcomes, a decision tree model was developed in Python using Google Colab. The dataset included grades from key academic indicators such as OJT, Capstone, GPA, Programming, Math, Ethics, and Communication. The trained model achieved an accuracy of 83.33%, effectively distinguishing patterns of academic risk. Specifically, students with Capstone grades of 4.00 or higher, or multiple failing grades in core subjects, were frequently classified as "At-Risk." These findings provide actionable insights for academic intervention, curriculum refinement, and program enhancement. The research supports evidence-based decision-making and contributes to Sustainable Development Goal 4 which is Quality Education by promoting inclusive and data-driven approaches to student success.