This study uses data from IMDb and TMDB to build two machine learning models. One model predicts movie ratings, and the other classifies movie genres. To predict ratings, we work with a carefully selected subset of IMDb data, including details like titles, genres, ratings, and crew roles, to create a structured dataset. We focus on Gradient Boosting Decision Trees (GBDT), including XGBoost and CatBoost models, and check their performance with metrics like Mean Squared Error (MSE) and Mean Absolute Error (MAE). Out of the tested models, the Gradient Boosting Regressor gives the best results, achieving a balance between speed and accuracy. For genre classification, we collect plot summaries from TMDB. Using Sentence Transformers, we create embeddings that capture the relationships between genres. We feed these embeddings into a convolutional neural network (CNN) that incorporates Conv2D layers with a 1x2 kernel size and MaxPooling2D with a 1x2 pool size. Results suggest that content management may optimize rating and genre prediction and provide insights.