Description: Recommender systems (RecSys) are the engine of the modern internet and play a critical role in driving user engagement, while helping users to find relevant items that match their preferences learned from their historical interactions on other items. In many recommendation domains such as news, e-commerce, and streaming video services, users might be untraceable/anonymous, their histories can be short and users can have rapidly changing tastes. Providing recommendations based purely on the interactions within the current session is an important and challenging problem. To address that, many methods have been proposed to leverage the sequence of interactions occurring during a session. In this talk, we present how transformers-based architectures provide accurate next-click predictions for the short user sequences. At the end of the talk we will showcase a demo of how to train and evaluate a transformers-based session-based recommender model using Merlin Transformers4Rec open-source library.