Personalization and Recommendations Platform for Omni-channel Commerce

I'll outline some practical lessons and challenges learned from training and deploying AI-powered learning systems that power personalization and recommendations for the world’s largest omni-channel retailer. How do we handle petabytes of omni-channel data to train recommender systems, model customer preferences and affinities amenable for online inference, and (importantly) how do we account for a multitude of micro intents that manifest in user-session logs? I'll cover the foundational elements in our personalized platform and offer a deep dive on the recommendation engine that powers the omni-channel repurchase journey. I will also highlight the advances NVIDIA GPUs and other software components, like NVIDIA Merlin and Triton Inference Server, have brought to our platform.