Abstract In the era of digital content consumption, personalized recommendations have become essential for enhancing user experience and engagement. This report presents the development of a personalized movie recommendation system using collaborative filtering techniques implemented in C++. The system predicts user ratings for movies based on historical preferences and similarities with other users. The recommendation process involves loading a user-movie ratings matrix from a CSV file, calculating user similarities, predicting ratings for unrated movies, and generating a ranked list of recommended movies. The system was tested with a dataset containing ratings for 20 movies from 20 users. The results demonstrate the effectiveness of collaborative filtering in providing personalized movie recommendations, thereby improving user satisfaction and engagement.