The back squat is recognized as a highly effective exercise for enhancing athletic performance, requiring the coordinated engagement of various muscle groups and strengthening key movers essential for explosive movements like jumping, running, and lifting. Individuals who are new or inexperienced in fitness require precise training regimens to enhance their techniques and reduce the risk of injuries, rather than investing in costly coaching services. Utilizing programs and applications designed for home workouts proves to be a fitting alternative in such cases. This study presents a markerless methodology that measures the angle between the thigh and the vertical during the squat movement to provide visual insight and assessment for the user. We verify it by collecting actual data and comparing it with the facts estimated by the model. The angle between the thigh and the vertical was determined using the markerless pose estimation model “MediaPipe Pose” in combination with Computer Vision (CV) techniques. To assess the accuracy, the calculated angles were compared with reference data for each video frame. 30 data samples from 3 people of different heights were collected, the average absolute value MAE was 4.3°. The study demonstrates that the angle between the thigh and the vertical can be precisely measured using a machine learning-based solution for posture monitoring. This markerless approach offers a reliable alternative for professionals in biomechanics, sports medicine, physical therapy, and other medical fields to assess thigh-to-vertical angles during squatting movements, and other researchers can integrate this method into their practical application solutions.