NODES AI 2026 - Beyond Recall: Creating Agentic Personas with Persistent, Evolving Memory

This talk features Mohamed Fazil at NODES AI, focusing on the creation of agentic personas with persistent, evolving memory in AI agents. The session emphasizes the limitations of current stateless AI assistants that struggle with maintaining context and building trust over time. Participants will learn how to leverage Neo4j as a long-term memory layer, allowing AI agents to develop and adapt their personas based on user interactions. The talk will cover practical applications, including designing a graph schema for agent memory and integrating the LLM with Neo4j to enhance recall and self-reflection abilities. By the end of the session, attendees will gain insights into transitioning from ephemeral chat histories to robust, graph-native memory architectures.