Filmed at https://2017.dotai.io on April 25th in Paris. More talks on https://dotconferences.com/talks Recommender Systems are arguably the most common business application of Machine Learning systems. Recently a new blend of recommender systems have been developed by leveraging the tools and the modeling flexibility from the Deep Learning ecosystem. This presentation gives an overview of the main RecSys concepts such as matrix completion for collaborative filtering and relate those to current trends in Neural Network architectures.