Collaborative filtering is an approach used by recommender systems to provide personalized recommendations based on user interactions, such as ratings. This method allows users to "collaborate" with each other through implicit feedback, enabling the system to infer what a user might like. Collaborative filtering has been around for a long time and was one of the first approaches used in building recommender systems. It relies on user interactions, such as ratings, likes, and dislikes, to make recommendations. The algorithm uses Singular Value Decomposition (SVD) to decompose the user-item interaction matrix into two constituent matrices: user features and item features. These latent features can be used to predict a user's rating for a movie they haven't seen before. Collaborative filtering has several advantages, including its ability to handle large datasets and provide personalized recommendations. However, it also has some disadvantages, such as requiring large amounts of data and being sensitive to outliers. With Redis and RedisVL, it is possible to build a highly scalable and customizable collaborative filtering recommendation system that can be used in various industries.