Conventional solar tracking systems usually adjust the position of the solar panel based on position of the sun in the sky excluding other factors which may affect the power production. This system implements a new method for tracking solar position utilizing machine learning algorithms to estimate the best tilt angle with respect to extra climatic parameters including temperature, relative humidity, and GHI. The developed system also has a built-in feature of evaluating and actually applying different kinds of Machine Learning models such as the Feedforward Neural Network (FNN) and Random Forest that can improve the tilting angle forecast and further optimize the output power. Furthermore, it integrates a Maximum Power Point Tracking (MPPT) algorithm, which helps in dynamically varying the duty ratio of a boost converter to help in stabilizing and optimizing the power delivery. The performance of the proposed system is substantiated via accurate data analysis and power estimation techniques along with vast enhancements over more conventional solutions with respect to power and efficiency.