Just-in-Time Feedback for Adaptive AI Agents

Welcome to the AI research bites. This series of short and informative talks showcases cutting-edge research work from ServiceNow AI Research team. The AI Research Bites are open to all, especially those interested in keeping up with the fast-paced AI research community.

AI agents often fail not due to lack of intelligence, but because small mistakes compound over long horizons. In this presentation, Hadi Nekoei introduces JEF HINTER, a memory-based framework that distils offline experience into compact, reusable hints. By zooming in on decisive moments and reflecting them into actionable guidance, agents adapt without retraining. This enables transparent, efficient, and scalable improvement from past successes—and failures.

Paper: https://arxiv.org/pdf/2510.04373v1

ServiceNow AI Research team: https://www.servicenow.com/research/