Heart disease remains a leading cause of mortality globally, necessitating early diagnosis for effective treatment. Traditional heart disease prediction models, while useful, often suffer from inefficiencies in handling structured tabular data, high computational cost, and poor generalization. This study proposes a hybrid machine learning model combining Extreme Gradient Boosting (XGBoost) and Stacking Classifier to address these limitations. XGBoost excels in efficiently handling feature interactions and missing data, while Stacking Classifier enhances model performance through ensemble learning. By leveraging a variety of heart disease risk factors, including age, cholesterol levels, blood pressure, and ECG results, the proposed system offers a faster, more reliable, and scalable solution for heart disease prediction. This hybrid model aims to improve prediction accuracy, reduce training time, and enhance generalization, providing a robust tool for healthcare professionals to make informed, timely diagnoses. Keywords: Heart Disease, Machine Learning, XGBoost, Stacking Classifier, Prediction Accuracy, Healthcare, Ensemble Learning, Feature Optimization.