In today’s digital era, users are overwhelmed with abundant content across various streaming platforms. A personalized recommendation system has become essential for enhancing user experience by helping individuals discover movies that align with their preferences. This paper focuses on developing a movie recommendation system utilizing collaborative filtering, a widely adopted approach that makes recommendations based on user behavior and preferences. Specifically, we implement a user-based collaborative filtering model where the system predicts a user’s potential interest in a movie by analyzing similar users’ preferences. The model leverages Singular Value Decomposition (SVD) to reduce dimensionality and improve the efficiency of similarity computations. Additionally, a randomized search is employed to optimize hyperparameters, further improving the accuracy of the predictions. We implemented SVD in addition to a randomized search on Movielens 100K and 1M dataset, that performs 10% higher than the existing model. This approach enables the recommendation system to provide personalized movie suggestions, contributing to an enhanced user experience. By employing collaborative filtering, our system demonstrates how data-driven solutions can transform vast user interactions into actionable insights, offering a seamless and intuitive movie discovery experience.