This thesis focuses on the development of a predictive model for the interfacial tension (IFT) between carbon dioxide (CO2) and brine using advanced machine learning techniques. Accurate prediction of IFT is crucial for optimizing processes in carbon capture and storage (CCS) and enhanced oil recovery (EOR), both of which are vital for energy production and environmental management. Traditional experimental methods to measure IFT, while precise, are often expensive and time-consuming. This research aims to create a robust and efficient regression model to predict IFT, thereby reducing the reliance on extensive experimental procedures. Utilizing data from extensive literature reviews and experimental studies, this thesis develops and compares several machine learning models including Linear Regression, Multi-Layer Perceptron (MLP), Decision Tree Regressor, and K-Nearest Neighbors (KNN). The models are developed using data that includes various temperatures, pressures, salinity levels, and types of salts. Their performance is assessed using metrics like the coefficient of determination (R²) and Mean Absolute Error (MAE), with repeated K-fold cross-validation applied to ensure the models are robust and can generalize well. The results indicate that machine learning models, particularly the Decision Tree Regressor and Multi-Layer Perceptron (MLP), offer high predictive accuracy for CO2 and brine IFT across varied conditions. This thesis also includes a comparative analysis of the best-performing models against existing empirical correlations and experimental data, demonstrating significant improvements in prediction accuracy. By integrating machine learning techniques with traditional experimental data, this research contributes to the field of fluid dynamics, offering a cost-effective and reliable method for predicting IFT.