The use of Agile approach is on the rise in software engineering projects pertaining to data science, ML, and AI. But research on the actual operation of such programmes is severely lacking. The need to adapt to new technologies has increased in recent years, and many academics and software engineers in the software industry have taken an interest in the agile development approach as a result. The proposed work in this study deals with the study and analysis of the most popular techniques used in every category of estimation practice used in agile development. Further, the machine learning techniques based on regression, and a few machine learning techniques like Random forest Regressor and Polynomial Regression are applied and the results are observed and analyzed for the estimation accuracy in terms of MSE and R-Square that get 98 and 97%, respectively. Also, have offered some recommendations for how ML teams and projects might benefit from and make better use of Agile.