<p>Slope stability analysis is a critical aspect of geotechnical engineering with significant implications for infrastructure safety and disaster mitigation. This study presents a framework that integrates the limit equilibrium method (LEM) with machine learning (ML) techniques to enhance the accuracy, computational efficiency, and adaptability of slope stability assessments. The methodology begins with a comprehensive dataset of geotechnical and geometrical parameters that undergo rigorous pre-processing to ensure robustness and reliability. Sensitivity analysis was performed to quantify the influence of the key parameters on the safety factor (<i>F</i>_Slope). This study evaluates various ML models, including bagging methods such as Random Forest (RF) and Extra Trees (XT), and boosting algorithms such as Adaptive Boosting (AdaBoost), Gradient Boosting (GBoost), CatBoost (Categorical Boosting), and XGBoost (eXtreme Gradient Boosting). GBoost emerged as the top-performing model, achieving an exceptional R² of 0.998, weighted mean absolute percentage error of 0.011, and root mean square error of 0.018. Surrogate modeling is employed to further enhance the computational efficiency while maintaining the predictive accuracy. The developed framework was validated through real-world case studies, thereby demonstrating its practical applicability across diverse geotechnical scenarios. By combining the theoretical rigor of LEM with the adaptability of ML and surrogate models, this study provides a transformative approach for slope stability analysis, offering a balance between computational efficiency and predictive reliability. The results highlight the potential of this hybrid framework to revolutionize geotechnical engineering practices, enabling safer and more efficient slope stability assessments in critical infrastructure projects and landslide-prone areas.</p>