<p>This research presents a user-friendly Python tool to automate single pile settlement predictions, making advanced machine learning (<i>ML</i>) techniques more accessible to geotechnical experts. Leveraging a comprehensive dataset of 656 records from 41 full-scale bored pile load tests conducted under various Egyptian subsoil conditions, we rigorously trained and evaluated six prominent <i>ML</i> models: Gaussian Process Regression (<i>GPR</i>), Extreme Gradient Boosting (<i>XGBoost</i>), Gradient Boosting Machine (<i>GBM</i>), Random Forest (<i>RF</i>), K-Nearest Neighbors (<i>KNN</i>), and Support Vector Regression (<i>SVR</i>). Among these, <i>GPR</i> emerged as the top-performing model, showcasing exceptional predictive accuracy and robustness, evidenced by consistently high coefficient of determination values and low error metrics on unseen test data, as well as tight clustering of predicted versus actual settlement values. A key feature of this study was the integration of Monte Carlo simulations to quantify uncertainties associated with input parameters. Results were visually represented through load-settlement curves with 95% confidence intervals, providing a comprehensive assessment of prediction reliability. Furthermore, a detailed <i>SHAP</i> feature importance analysis identified the most influential factors in the <i>GPR</i> model’s predictions, aligning with established geotechnical principles. Finally, this work offers a reliable and efficient framework for forecasting single pile load-settlement behavior, enhancing the accuracy and speed of geotechnical analysis and contributing to the development of more dependable prediction tools for civil engineering infrastructure.</p>