This research paper aims to explore the feasibility and potential of measuring and predicting student’s academic performance using machine learning by analysing the extensive student datasets. We aim to proactively identify students at risk of underperforming, facilitating timely interventions. This paper delves into predicting student performance using regression models and ML methodologies. In an era dominated by data-driven insights, the academic sector is no exception. Our methods are based on Mean Square Error, Root Mean Squared Error and R-squared metrics. The findings aim to Bolster Academic approaches, potentially offering insights to amplify student outcomes. A comprehensive architecture considering aspects of scalability, latency and security has been established.