To build scalable and reliable AI agents across enterprise data, agentic workflows need a knowledge layer. Join us at AI Dev NYC at booth 4 in and we'll show you how Neo4j provides AI with connected context, long-term memory, and traceable reasoning for explainable, agentic decision-making and answers.
Stephen Chin, VP of Developer Relations
Even a galactic empire can fail when its AI retrieves the right documents but misses the connections between them. Following one ambitious stormtrooper through a series of increasingly consequential knowledge failures, this talk explores how GraphRAG, graph-based retrieval, and graph memory have become a cornerstone of modern enterprise knowledge systems.
Based on the O’Reilly book GraphRAG: The Definitive Guide by Stephen Chin, Michael Hunger, and Jesús Barrasa, the session provides a practical map of the patterns behind these systems. We begin with the limits of vector-only RAG when questions are connected, multi-hop, dynamic, or require an explainable evidence chain. From there, we cover knowledge graph construction from structured and unstructured data, ontology-guided extraction, entity resolution, and the roles of lexical, domain, and meta graphs. We examine hybrid retrieval using vector search, BM25, graph filtering, guided traversal, Cypher templates, Text2Cypher, graph embeddings, community detection, and query-focused summarization. We then connect retrieval to graph memory and context graphs that preserve conversations, reasoning traces, tool activity, reusable procedures, and changing operational state.
Along the way, we address evaluation, provenance, access control, freshness, governance, and the trade-offs involved in moving these systems into production. Attendees will leave with a decision framework for designing enterprise knowledge systems that are accurate, explainable, adaptive, and capable of supporting agents in large-scale enterprise scenarios.
Nyah Macklin, Senior Developer Advocate, Neo4j
Jennifer Reif, Senior Developer Advocate, Neo4j
Your agent just gave a customer the wrong refund decision and attached a real company policy to prove it. Now the customer has a screenshot. The passage matched the question. It came from a policy version that didn’t apply to that customer on that date.
In this hands-on workshop you’ll pull apart how that happened and repair it with the simplest approach that works. Then you’ll take on a harder case, where the answer depends on connections across customer, product and policy data. A small set of checks shows when the system has found applicable evidence, and when it should stop.
Hosted by Neo4j - This special event brings together curious young minds and their families for a fun, hands-on introduction to artificial intelligence, coding, and technology.
Â
​Guided by Cassandra Chin, participants through interactive sessions designed to spark creativity, build confidence, and make complex ideas accessible and fun. Parents are encouraged to join.
Â
Sunday, November 29, 12:30–4:00 PM ET at Jay Suites Chelsea, 159 W 25th St, New York. Free; registration required.
Â
Speaker: Cassandra Chin, Author & CS student
Â
The challenge: "Context, connected"
Join the Relate by Neo4j Hackathon, a month-long online event celebrating the launch of Relate by Neo4j.
Build a project that shows how connected data helps agents, apps, or people find and use the relationships that matter. Your project can be new or extend something you already have. Share your code and documentation so others can learn from and reuse your work.
Relate is an open-source embedded database for AI agents and the engineers who build them. It runs as a library inside your application, with no signup or server to deploy.
