Neo4j

What graph embeddings capture? a node and everything around it, compressed into a vector. Getting started used to mean choosing models and tuning parameters. Mahdi Karabiben’s post introduces a new Embeddings API for Aura Graph Analytics, now in preview. It’s one line. https://bit.ly/4yFlNAt

No more guessing for your #AI: we are announcing Neo4j Context Engine. It 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. Currently in private preview. https://bit.ly/4zyWC3a

Vector Graph or GraphRAG? GraphRAG outperforms vector-only RAG, BUT it does not replace it - Take a look at this blog and find out whY: https://bit.ly/4ynCCzu

Agentforce owns the conversation. Neo4j #AuraAgent owns the graph. One sends a question over MCP, the other traverses the data and returns a grounded answer. No Cypher on the Salesforce side. https://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! https://bit.ly/4zoZ6RX

"Across two deliberately hostile catalogs (278 Census tables on BigQuery and 264 legacy tables on Databricks), a #Neocarta semantic layer in Neo4j makes Text2SQL agents faster, cheaper, and more accurate." claims Mounir Babari. Full benchmark? https://bit.ly/4APuw5a #semantics #Neo4j #AI

If you don't know what a graph visualization is, stay here: 🌸 Neo4j Bloom 📋 Neo4j Dashboards 📚 Neo4j Visualization Library Take a deep dive into #GraphVisualization, how it helps your workflow, and how to use it. A blog to share and save :) https://bit.ly/4hJywf7

Working with Dates and Durations: Use Neo4j temporal types to filter, compute, and search date-based data https://bit.ly/3VYz7lC Check out this course.

Getting an #AI application into production? It's only one part of the journey. We know challenges shift right away. One graph behind one application becomes dozens of workloads across teams, tenants, and environments. Check out #Neo4j capabilities that address #AI scalability. https://bit.ly/4yg1sSR

This guide breaks down what a decision trace actually records: its outcome, each reasoning step, every tool call, and its result, and links to the entities and conversations the decision drew on. All of it is stored in a context graph. https://bit.ly/46t2xuh

We really feel you dont want to miss this workshop, and you should invite your friends to it! #sharingisCaring Discover why AI agents do not hallucinate because of bad prompts but because of bad architecture. https://bit.ly/4xleahN

Understand centrality algorithms in 2 hours with #GraphAcademy! Six loose families, and the algorithms inside a family answer similar questions. Which family is the right one for an analysis? https://bit.ly/4rgNzkP

Today, we are introducing Neo4j GraphAware Financial Crime Intelligence - one connected, graph-native foundation for modern financial crime operations. Details and demo: https://bit.ly/4AgLzwU

Security is the part most teams skip when building agentic AI. Today, we show you how to turn your graph into a policy engine your agents can actually trust - covering #MCP, #OAuth2, and the new #AuthZEN standard. https://youtube.com/live/6k096_wfHCY #agenticai #knowledgegraph @indykite.bsky.social

Define and enforce a graph schema using graph types in #Neo4j! https://bit.ly/4xRxe8y 6 lessons across 1 module-start today!

Your agent can query the knowledge graph. Can it read the CEO's salary? Delete an asset with zero trace? Break GDPR without anyone knowing? #RBAC won't save you. Your #graph can. Join us next week: https://youtube.com/live/6k096_wfHCY #Neo4j #AgenticAI @indykite.bsky.social

How graph databases can supercharge your knowledge base? Zach explores Karpathy’s LLM wiki recipe, showcases ki (open source), and shares research from Newcastle University proving agents achieve 2x better precision/recall and major gains in truthfulness and relevancy. https://bit.ly/4ylrBza

Weekend of building! Got Aura credits? Save this guide on how to start building 🚀 Install the agent skills Build something in one sitting Size your instance - don't guess! ;) https://bit.ly/4qLXHBC #Neo4jAura #AuraDB

1operating structure - Business units, teams, roles, reporting lines, modeled in #Neo4j.at go deeper 1st one? Operating structure - Business units, teams, roles, reporting lines, modeled in #Neo4j. This layer answers organizational questions. https://bit.ly/4gnNlV4

A practical, engineering-focused approach to building a semantic layer on Neo4j with ontology at the core, with Jesús Barrasa. Webinar on demand: https://bit.ly/4frB8NB #ontologies #graphdatabase #AI #semantics

Vector + graph RAG delivers 2x the factual accuracy of vector alone. Have you seen the latest research conducted by The National Innovation Centre of Data (NICD)? @jimwebber.org https://bit.ly/4qyNuZd

Aura Graph Analytics is on-demand, session-based graph compute: pay-per-minute, isolated from your live database, and able to read graphs from any data source. What AGA is, when to reach for it, and how to drive sessions from any #Neo4j driver. 1-hour course! https://bit.ly/4qqgRNi

Jeremy Morgan mapped 10 SQL queries to their #Cypher equivalents, from SELECT to recursive traversal, using a real npm dataset you can run yourself. https://bit.ly/4xzEWn7 blog

Bloom can now run graph algorithms across an entire database, not just the current Scene. Jeff Gagnon walks through the difference with a concrete example, showing how whole-graph results can tell a very different story than a local one. https://bit.ly/45mMPAw

Today, see how graph agents extract multimodal clinical entities, traverse disease hierarchies, and combine embeddings, BM25, and graph traversal in one pipeline - built on FHIR and the WHOLE framework. https://youtube.com/live/8otCxnotd2E #Neo4j #GraphRAG #HealthTech #AIAgents

Jeremy asks himself, "Is Max Verstappen connected to Juan Manuel Fangio? Could a driver who retired in 1958, decades before Max was born, connect to him through a chain of teammates? How many links does it take?" Are graphs showcasing something cool here? Check it out https://bit.ly/4z3Jnbj

Is this time of the year when we start talking about the most expected annual conference dedicated to engineering better intelligence... #NODES2026! We have a day: November 12 NODES is free, global, and for all: experienced or beginners. Join :) https://bit.ly/4pXd3Tq

Patient data lives in scans, labs, and notes. Most AI reads each one alone - and context disappears between encounters. Let's fix that with graph-based context memory: multimodal entity extraction and hybrid retrieval https://youtube.com/live/8otCxnotd2E #Neo4j #GraphRAG #HealthTech #AIAgents

What is a #ContextGraph? Why do #AI agents need a context graph? Fundamental questions for accurate agents. Find out more: https://bit.ly/4xkj6nx

Graph-native agent memory for the .NET world — built from scratch, and verified against Neo4j’s own Test Compatibility Kit. https://bit.ly/3ROH9vF by José L. Latorre and Zaid Zaim

Research by the UK’s National Innovation Centre for Data shows that graphs can significantly outperform vector-only LLM systems in accuracy and cost. @itpro.com writes: www.itpro.com/business/dat...

Will Lyon on August 6: how context graphs unify short-term messages, long-term entities, and reasoning traces in one queryable structure, and how successful execution patterns get distilled into reusable SKILL .md artifacts new agents can use from day one. https://bit.ly/4yWei9z

AI on Your Lakehouse: Context Comes in Shapes, Not Queries - #Announcement New #GraphAcademy course release 📢 (right on time for the weekend!) 17 lessons in 1 HOUR. Details: https://bit.ly/4wkp6wo

Tune in today and see #supplychain impact analysis done right: find which vehicles are affected, whether a substitute clears compliance or where supplier concentration is quietly creating risk. https://youtube.com/live/WAY-r9SweRE #graphdatabse #agenticai

Wonderful news! Virtual Graph is now available in the Aura Console for Snowflake, Databricks, and Google BigQuery. -Connect a source, generate a model, and run your first Cypher query in minutes. - - Public preview is free to use. Enjoy! https://bit.ly/3Rp05RK

Use #Neocarta to easily build semantic layer graphs for your agentic applications in #Neo4j Worth reading if you are building agents on Google Cloud and hitting the limits of flat schema files: https://bit.ly/4vepyvP

A chip goes dark. Vector search finds lookalikes. But can it tell you which vehicles are affected? Whether the substitute clears compliance? Where your supplier concentration creates risk? Graphs can! https://youtube.com/live/WAY-r9SweRE #graphdatabase #supplychain #agenticai

Today: We built a network intrusion detector using Neo4j Graph Analytics for Snowflake: KNN to connect similar attack events, GraphSAGE to classify malicious traffic https://youtube.com/live/4Iq3iU68Nqk #snowflake #cybersecurity

Today, we are announcing MCP for Aura, a hosted MCP server built into your Aura instance. MCP for Aura eliminates the tax by providing a fully hosted, native MCP server within Neo4j Aura, allowing developers to connect AI agents to their graph data in minutes. https://bit.ly/4b63qf7

Get started on #Neo4j: ✔️ Start for Free ✔️Deploy via Cloud Marketplaces ✔️Scale on Demand ✔️Import Your Data Fast What's the best plan for you? ⬇️ (special offer for #startups - check it out: https://bit.ly/4xwg5S5

SQL shouldn't be able to do this. But it can. 🤯We built a network intrusion detector using Neo4j Graph Analytics for Snowflake All natively in Snowpark. https://youtube.com/live/4Iq3iU68Nqk #snowflake #cybersecurity

Say hello to our refreshed #GraphAcademy landing page with E.L.A.I.N.E, our AI learning assistant, to the front and center, to help you find your next course and guide you through your learning. 🚀 https://graphacademy.neo4j.com/

Neo4j Enterprise Studio is here: one self-managed workbench combining Query, Bloom, and Dashboards for teams working with graph data. Write and debug Cypher. Explore data without code. Build and share interactive dashboards. All in one place, on infrastructure you control. https://bit.ly/4eLGQtd

Most AI agents lose context mid-task. A conversation buffer and a static knowledge base are not enough. Jim Webber on why context graphs give agents three connected memory layers that actually last. neo4j.com/blog/agentic-ai/context-graph-ai-agent-memory/

Most #AIAgents are impressive in conversation and isolated in operation. ⏬ Why context graphs fix this with three connected memory layers: long-term knowledge, short-term conversation history, and reasoning traces that persist across sessions. https://bit.ly/4wRE2ST

Don't miss out today: #goingmeta and we're going deep on ontologies and context graphs! We are looking forward to the conversation with you! https://youtube.com/live/JGFoTQt4GA0

Paulien van Alst @paulienvanalst.bsky.social - Thinking in Relationships: Practical Graph Database Modelling in Java with Neo4j Auf dem Java Forum Nord 2026: javaforumnord.de/2026

Next week: Going Meta and we're unpacking ontologies and context graphs - the layer that turns your data into something AI can reason over. https://youtube.com/live/JGFoTQt4GA0 #ontology #contexctgraph #knowledgegraph

Our VP of Product Marketing flies his own amateur-built airplane. And built a digital twin of it on Neo4j AuraDB. 1,500 nodes. 3,800 relationships. He just asks it things from the hangar. Full story: bit.ly/4wbsEAz https://bit.ly/4wbsEAz

Tune in today and see a full audit trail in Neo4j in action! https://youtube.com/live/Wb_UOg93BLM #neo4j #aigovernance #agenticai