In today’s world, physical fitness and exercise holds great importance as it helps improve the quality of life. However, incorrect postures during exercises may lead to severe long-term injuries like Tendinitis or Hamstring strain. Hence, we propose an application that will analyze the posture of the user performing a particular workout session and comparing it to an online video or a reference video and offering suggestions to improve it if necessary. Our system will provide visual feedback while performing a workout by synchronizing user and reference videos. We will analyze the angles between the limbs of the body and compare it to the reference video to provide corrective action to the user using PoseNet on TensorFlow. With the help of this model, fitness enthusiasts can perform a particular workout accurately without getting injured and with proper guidance.