This project aims to build a restaurant recommender system using content-based and collaborative filtering techniques, using the Yelp dataset. Work involved handling large JSON datasets, engineering features from nested attributes, and experimenting with cosine similarity and transformer-based NLP for review analysis. Key learnings included managing compute constraints, avoiding exploration paralysis, and understanding the limitations of basic similarity methods. The project is currently on hold but will be resumed soon to explore collaborative filtering and hybrid approaches further.