This study uses computational methods to investigate the fluid flow characteristics around a wavy NACA 63(4)-021 hydrofoil near the water surface. It introduces notable contributions to the literature on sinusoidal edge hydrofoils by examining the behavior of this specific hydrofoil configuration close to the water surface, a previously unexplored aspect. Furthermore, the study presents a comprehensive review and analysis of various machine learning (ML) techniques applied to computational fluid dynamics (CFD) and experimental data, offering a novel approach in this field. The three-dimensional Reynolds-averaged Navier–Stokes (RANS) equations are solved using an implicit finite volume approach to simulate the turbulent flow around the hydrofoil near the free surface. The realizable k–ε turbulence model accounts for turbulent flow effects at varying submergence depths. Four machine learning models in Python are developed to predict lift and drag coefficients. Evaluation of these models on training and test datasets reveals that xgboost achieves the highest accuracy, with an impressive R2 score of 0.9775 on the test dataset. Therefore, xgboost is recommended as the optimal model for future applications in this context. The study's findings are presented for the wavy hydrofoil at three submergence depths and three angles of attack (AOA). Key results include wave profiles, magnitudes, total pressure contours around the hydrofoil and free surface, and pressure, lift, and drag coefficients. The accuracy of the numerical simulations is validated by comparing the results with available experimental data, demonstrating good agreement between the two.