Summary
This blog post discusses Grab's development of a foundation model for AI-driven personalisation in their superapp, focusing on leveraging user, merchant, and driver data effectively. Key challenges such as integrating diverse data types, ensuring scalability, and developing a robust architecture are highlighted. By creating user embeddings that consider both long-term behaviors and short-term intents, Grab aims to enhance user experiences across its services, ultimately paving the way for a more intelligent, responsive platform. The model represents a shift towards a unified intelligence layer that can support various applications like fraud detection and personalized recommendations.