Learn with Neo4j's New "Get to Know Graph & GenAI" Webinar Series >>

Neo4j logo

Tracing the LLM Behaviour: Debugging and Evolving Agent Workflows with Neo4j

Session Track: AI Engineering

Session Time:

Session description

Modern LLM apps don’t just answer prompts—they reason, plan, and act across complex workflows. As these systems grow in capability, they also grow in opacity. When something goes wrong (the agent loops, misfires a tool, or takes an unexpected detour), developers are left in the dark. In this talk, we introduce a powerful method for tracing, explaining, and evolving LLM agent behavior using Neo4j. You’ll learn how to: - Model agent execution as a graph of memory hops, tool calls, intermediate thoughts, and transitions - Use Cypher queries to trace failures, detect reasoning loops, and isolate ambiguous decision paths - Apply graph metrics to evaluate agent complexity and fragility - Visualize and replay agent reasoning for debugging and auditability We’ll demonstrate how Neo4j transforms opaque runs into explainable, inspectable agent flows, enabling developers to ship smarter, safer, and more trustworthy GenAI systems.

Speakers

photo of Rangesh Sripathi

Rangesh Sripathi

Principal Engineer, Verizon

Architect at Verizon

photo of Aayushi Sinha

Aayushi Sinha

Senior Engineer, Verizon (India)

Aayushi Sinha currently works as a data scientist for Verizon with eight years experience in the fraud detection domain. She is passionate about the new trends in the field of GenAI and keeps up with the latest models and their use cases in the market. Aayushi has also worked in application development as a full stack developer using Java and operated and managed cloud infrastructure by AWS.

photo of Sridharan Sundaram

Sridharan Sundaram

Data Science Engineer, Verizon (India)

Sridharan Sundaram is an engineer in data science with more than four years of experience in fraud analytics. He specializes in data science, graph analytics, Cypher, Python, and ML. Holding an M.Tech in Data Science, Sridhar is passionate about AI, graph networks, predictive modeling, and LLMs, to name a few.