This Week in Neo4j: NODES AI, Agents, Cypher, AI Memory and more

Senior Developer Marketing Manager

April 24, 2026

4 min read

Welcome to This Week in Neo4j, your fix for news from the world of graph databases!

I am still full of excitement from NODES AI last week! The recordings are now live, including an extended keynote cut with bonus content that didn’t make it into the event.
At NODES AI, we also launched an Aura Agent Hackathon: Win $$$ credits and prizes and bring your idea to life with Aura Agents!

In this edition, we are also learning how to wire Neo4j into Gemini Enterprise as a production-grade graph engine, understanding why native Cypher is ready to replace APOC, and exploring how context graphs give AI agents not just memory but the ability to reason across it.

Share your experiences and influence the future of Neo4j products: Join the Neo4j User Research panel! It’s a chance to connect directly with product development teams, get paid compensation, hear about what we are working on and more!

Happy Graphing,

Alexander Erdl

COMING UP!

Livestream: The Knowledge Layer: How Neo4j Makes AI Agents Production-Ready on May 6
Conferences: Find us at AI Dev 26, San Francisco on April 28-29, AgentCon, San Francisco on May 4, AI Agent Conference, New York City on May 4-5, Knowledge Graph Conference, New York City on May 4-8, Data Innovation Summit, Stockholm on May 6-8 & Devworld Conference, Amsterdam on May 7-8
Meetup: Meet us in Nuremberg, DE & London, UK on May 7
All Neo4j Events: Webinars and More
GraphSummit Series: Transform Your Enterprise with Graph and GenAI – Next Stop: San Francisco on July 23

FEATURED COMMUNITY MEMBERS: Animesh Koratana & Lasse Andresen

Animesh Koratana is the founder and CEO of PlayerZero, an AI production engineering platform that autonomously resolves incidents and catches defects before release.
Lasse Andresen is the founder and CEO of IndyKite, building the identity and context layer that lets AI agents navigate enterprise data graphs with precision and governed access.

Together with Emil Eifrem, Will Lyon and Philip Rathle, they were guests at the Opening Keynote from NODES AI: “Exploring Context Graphs: From Data to Decisions”, where they explored how context graphs are emerging as a new foundation for AI systems that can reason, learn from experience, and improve over time. An extended cut is now available on the NODES AI Youtube Playlist.

NODES AI: Recordings Now Available

NODES AI was an amazing event last week with tons of interesting content around AI Memory, Context & Knowledge Graph and more. We are starting to add the event recordings to our YouTube playlist so you can watch any session you might have missed during the day.
The NODES AI opening keynote also got an extended cut with about 20 minutes more content we didn’t have time during the event!

AGENTS: Building a Neo4j Graph Agent for Gemini Enterprise

Kaustubh Darekar shows how to bring a graph agent to production on Google Cloud. The architecture uses MCP, Google’s Agent Development Kit and the Agent-to-Agent protocol to wire Neo4j into Gemini Enterprise as a fully decoupled, Cloud Run-hosted service, with OAuth 2.0 validation and per-user token tracking built in from the start.

CYPHER: Batching Like a Pro

If you’re still reaching for apoc.periodic.iterate to handle bulk updates in Cypher, Gemma Lamont makes a compelling case for switching to native CALL { … } IN TRANSACTIONS. Side-by-side, she compares readability, query planning, memory safety and error handling. Particularly around OOM risks and the awkward node-rebinding workarounds APOC requires, the differences become very clear.

AI MEMORY: From recall to reasoning: How context graphs upgrade an agent’s brain

Niels de Jong uses a surprisingly compelling metaphor – a hungry sheep navigating a forest full of wolves – to illustrate why most AI agents hit a ceiling: they can recall facts, but they can’t reason about them. The post walks through three levels of agent memory, showing how a context graph bridges the gap by connecting not just what happened, but why.

CONTINUOUS LEARNING

GraphAcademy: Learn how to create your own MCP server with “Building GraphRAG Python MCP tools“
Learn on Your Schedule: Go deeper into graph technology on Neo4j’s On-Demand webinar library
Workshops: Join our virtual classrooms workshops from Fundamentals to GenAI
New Webinar: How Context Graphs Power the Next Generation of Autonomous Agents – Americas, Europe, Middle East & Africa, Asia Pacific