Recent studies have increasingly focused on enhancing the accuracy of software project effort estimation within the Agile methodology framework. This trend emphasizes the integration of advanced machine learning and deep learning techniques, including neural networks and convolutional neural networks (CNNs). Additionally, optimization strategies—particularly in the early stages of modeling—have gained attention as a means to improve prediction outcomes. A central theme across these studies is the use of Story Points as a core metric for estimating software development effort. This research aims to explore and synthesize a selection of recent scholarly works that contribute to this evolving area, examining their methodologies, datasets, algorithms, and key findings.