No more guessing for your #AI: Today, we are announcing Neo4j Context Engine. Neo4j Context Engine enables enterprises to define, connect, govern, and continuously enrich business and data context, making it available to AI agents at the point of decision and action. Through our six-layer interconnected ontology, agents can give governed, reliable, and explainable answers. And the more you use it, the better it gets. Neo4j Context Engine is currently in private preview. Follow the link below to become an early design partner; bit.ly/4zyWC3a

AI energy at a packed @epignosishq today! 🚀 We’re building AI agents with memory alongside @adamcowley from @neo4j context graphs to connect conversations, knowledge and decisions. Hands-on learning, working agents, #JavaScript & #TypeScript. 💡 #AI #AIAgents #Neo4j

Which skill is your biggest gap right now? @AndrewYNg's team analyzed 10,000+ job postings to identify the skills that matter. Hear about what they found in the fireside chat with @emileifrem during #NODES2026. Join! The coolest online 24-hour conference with 100 speakers from around the globe. Join for free! bit.ly/4pXd3Tq

“In fraud detection, knowledge gives us the context to help understand not only what’s going on currently, but also how an entity has evolved their business transactions historically. That will reveal signals of fraud, waste, and abuse within the current data. That’s important because we’re trying to detect fraud proactively.” Thomas Luu, Director, Anti-Counterfeiting Team, Gilead Sciences Just released: The MIT Technology Review Insights survey of 300 senior data and technology executives to find out why only approximately 34% of agentic AI projects reach production. The answer isn’t missing data but the missing knowledge. The report you needed is here. Download today: bit.ly/4822whw #Neo4j #MIT #AIAgents #knowledgelayer #contextengineering

@neo4j Greetings to the Neo4j community. I am happy to announce that I have officially become a certified Neo4j Developer. 😀 I will be combining Neo4j with my JLNN. github.com/RadimKozl/JLNN

Vector Graph or GraphRAG? In a study by the UK’s National Innovation Centre for Data, GraphRAG scored roughly 80% higher on truthfulness than vector-only RAG. It answered 65.3% of 510 complex questions, against 28.9% for vector-only. It doesn’t replace vector search, though. For single-fact lookups and semantic matching, vector RAG is still the right tool. The blog shows how to tell which one you need: look at how your retrieval is failing, build a test set from your real traffic, and see whether GraphRAG moves the numbers. Happy weekend reading! bit.ly/4ynCCzu

Face recognition is a graph problem, not just vector search 🕸️ I’m speaking at #NODES2026 on building scalable identity systems with DeepFace + @neo4j Vector Search 🔹 Millisecond vector queries at scale 🔹 Centrality to fix false positives 🔹 Production-ready with minimal code

The Connected Startup, Episode 5: lcmd nitter.siddz.com/i/broadcasts/1OxwbnyyY…

Road to NODES 2026 - Build Your First Knowledge Layer nitter.siddz.com/i/broadcasts/1RKjpbnMN…

⚠️ Problem: One graph project turns into five. Each one needs dev, staging, and QA. Suddenly, a team is running twenty instances for what is really a handful of workloads. 🤩 Solution: Multiple databases are now generally available in Neo4j AuraDB Business Critical and Virtual Dedicated Cloud! Run multiple independent databases within a single AuraDB instance, sharing resources while maintaining full data separation. A simpler way to consolidate workloads while keeping databases separate and independently managed. How? bit.ly/4ycwaLF #AuraDB #Graphdatabases

Ask the right tool the right question 💪 Join Tomaz to learn how to build a hybrid RAG system with multiple retrievers that handle every question type, so you can ask the right tool the right question. This is a tech session where you'll learn to: ✔️ Build a production-ready agentic system with multiple retrievers from a sample dataset ✔️ Route semantic, computational, and relationship-based queries to the retriever built to answer them ✔️ Implement an answer critique function to check whether your original question is answered correctly Americas: bit.ly/4hHIFds Europe: bit.ly/4ifYUhV Asia Pacific: bit.ly/4s4dexw #AI #Graphdatabase #HybridRAG

Build a memory stack that lets your agent learn from its own execution traces. In this free, two-hour online workshop, you'll design a Neo4j schema that captures agentic execution traces, compose playbooks dynamically at query time, and build a Reflector→Curator loop that improves from both human corrections and real outcomes. 💡 The test case is value investing, with live demos and real code throughout. You also see where graph-based memory pays off, and where the trade-offs in latency and complexity start to matter. Build intelligent, graph-powered applications! bit.ly/4ywyP3U #NODES2026 #Neo4j #AI

Import a dataset into #Neo4j Aura, build a product recommendation dashboard with live filters and a graph visualization, publish it to a URL, and share it with stakeholders. The #Cypher queries that power it are provided. You just build. graphacademy.neo4j.com/cours… #Neo4j #GraphAcademy #AuraDB #NoCode #GraphDatabase

Building on IBM watsonx Orchestrate and want your agent to actually remember users across sessions? Kaustubh Darekar wired #Neo4j as both the knowledge layer and the memory layer: a native Orchestrate agent queries a companies graph through MCP with zero infrastructure, and an imported LangGraph agent adds cross-session memory through NAMS. Full architecture and code: bit.ly/4i5OZuj

Jakub Marchwicki on the cleanest multi-agent architecture for Salesforce and #Neo4j: Agentforce sends a natural-language question over MCP, Aura Agent resolves the customer, traverses purchase history, applies domain rules, and returns a grounded recommendation. Agentforce never touches #Cypher. Take a look on how to Ground Salesforce Agentforce with Neo4j — multi-agent setup over #MCP bit.ly/3TFipXS

What you were waiting for: turn your existing documents into rich knowledge graphs without writing a single line of code :) 🔜We are thrilled to share that the general availability for Document Intelligence, as part of Neo4j AuraDB, is coming soon! What changes with the general availability? ✔️ Larger collections and richer documents ✔️Build on the graph you already have ✔️More control through the interactive assistant Details on the release, capabilities, and what's next, here: bit.ly/4zoZ6RX #Neo4j #DocumentIntelligence

POV: before/after (a 5-minute time difference 🙃) Packed room at #GraphSummit London!