In the rapidly evolving field of sports technology, Artificial Intelligence (AI) is unlocking new possibilities for enhancing athletic performance. This paper introduces BadmintonTutor, an innovative web-based coaching application designed to assist beginners and novice badminton players in refining their skills. By leveraging advanced AI techniques such as pose estimation, object detection, and generative AI, BadmintonTutor delivers real-time audio feedback that helps players correct their posture and technique effectively. The system utilizes computer vision and machine learning algorithms to monitor and evaluate the biomechanics of badminton strokes and footwork, providing precise, actionable insights based on real-time video analysis. This continuous feedback loop not only enhances technique but also mitigates the risk of injury due to improper form. The development process, core components, and features of the application including motion capture, pose estimation, and feedback generation are thoroughly examined. Additionally, the study applies an adapted Technology Acceptance Model (TAM) Questionnaire to assess the system’s usability and effectiveness. With fully completed responses, the Cronbach’s alpha values indicate acceptable reliability for TAM questionnaire which is 0.868. The TAM questionnaire also demonstrates good content validity as confirmed by expert review. The implications of this research extend beyond badminton, offering a template for integrating AI-driven pose correction across various sports disciplines.