27 August 2025 14:30 - 15:00
Personalization at scale: Designing AI systems that learn and adapt
In a world where users expect systems to "just know" what they want, personalization has moved from a feature to a baseline. But building intelligent, adaptive systems that respond to evolving user behavior - without overwhelming infrastructure or compromising trust - is no small feat.
This session unpacks the end-to-end pipeline: capturing user signals, retrieving and ranking content in milliseconds, and orchestrating IoT devices - all while balancing real-time and batch inference, modeling the long tail, and preventing bias and fatigue.
No matter what kind of system you are building, you’ll leave with a blueprint for context-aware, fault-tolerant systems that continuously adapt and deliver intuitive, respectful experiences.