At its core, agents have models taking in information and deciding the next steps the agent should take. Currently, models are limited in the amount of context able to be retrieved, affecting the performance of agents running over enterprise-level chains. This session aims to highlight how to leverage graphs to store weights highlighting successful and failed agent executions. When in production, agents will perform RAG over this execution graph to retrieve successful executions as context to decide on the next steps, making enterprise-level agents reliable.
with Julian Saks
Get certified with GraphAcademy: https://dev.neo4j.com/learngraph
Neo4j AuraDB https://dev.neo4j.com/auradb
Knowledge Graph Builder https://dev.neo4j.com/KGBuilder
Neo4j GenAI https://dev.neo4j.com/graphrag