Increasing demand for accurate crop yield predictions in agriculture has been fueled by technological innovation. Machine Learning (ML) and Remote Sensing (RS) have become leading tools for precision and scalability in predictions. This review discusses the present state of the integration of ML and RS, bringing out methodologies, datasets, applications, and challenges in predicting crop yields. It has thus proven to show considerable promise for agricultural decision-making, resource use optimization, and improvement in food security through synergies between ML algorithms and RS data. The future trends and potential advancement are also discussed below.