Attention AI startups!! We have some big newsBig news for AI Startups! 🚀 We are now elevating our commitment by providing a dedicated team of Field Engineers focused on deep technical engagement. Brian O’Keefe wrote about what he learned during his few weeks leading this team. Check it out and apply to join our program today. bit.ly/42uItpv #Startups #AI #GraphIntelligence

Got my @neo4j cert a while back, but the Aura Agent Hackathon was the perfect excuse to finally build something cool with it. Full architecture breakdown live demo: community.neo4j.com/t/start-…

Learn graphs. Score points for your country. The World Cup inspired us with a cool competition on #GraphAcademy! Whether you care about ⚽ or not, it's time to learn and win points for your country! Step 1: Go to cup.graphacademy.neo4j.com and join your country Step 2: Complete courses and... earn points! 🏆You can win #Aura credits and Lego sets on every round! The tournament runs from June 11 to July 19. More than 40 courses are waiting for you. Get started 🚀

⏰ Few days left: #AuraAgent Hackathon Submissions Close June 15 at 11:59 PM PST Curious what other developers are building? ➡️ community.neo4j.com/invites/… The community has already reached 24 submitted projects, with developers building graph-powered AI agents across multiple spaces. From graph-powered healthcare assistants and cybersecurity tools to earnings call analysis, cricket intelligence, chess analytics, and geospatial supply chain optimization, the variety of projects being built is a great example of what happens when developers combine graphs and #AI. You are on time to build! • Complete the Aura Agent course • Claim your Aura Credits • Build a graph-powered AI agent • Submit your project • Share your work with the community Whether you're interested in knowledge graphs, #RAG, agentic workflows, recommendation systems, analytics, search, #cybersecurity, finance, healthcare, sports, or something entirely new, we'd love to see what you create. Join! community.neo4j.com/invites/…

🚀 Excited to have @neo4j supporting the launch of @RocketRideAI Cloud on June 18! Graph-native data the harness that ships #AI to production, together!!! 7 days to @RocketRideAI Cloud 🚀 📍 SHACK15, SF | 📅 June 18 🎟️ Save your seat: lnkd.in/gfNdcy6J 🔗 Tell us what you're building on our Discord: lnkd.in/gKZhS8ZW #ProductionAI #AILaunch #RocketRideCloud #AIDE

Great to see @steveonjava on stage at #vectorspaceday. Learned a lot about what can be achieved together with @qdrant_engine and @neo4j

🏐 🏆 We are all feeling the World Cup Spirit, right? Benjamin did a cool demo connecting Virtual Graph to our World Cup Data in Snowflake Check out the steps to go from table to graph in minutes and how to find hidden patterns among teammates in our data. Which club teams secretly power multiple national squads? Watch the full (and at a regular speed 🙃 ) here: bit.ly/4v7F1xC BTW, did you hear about Virtual Graphs? Virtual Graph translates Cypher to SQL and pushes queries to Databricks, Snowflake, and other SQL databases. No ETL. No Data movement. Take a look at the details here: bit.ly/43EdGH2 #VirtualGraph #AI #knowledgelayer

Our @Databricks Summit plans for June 15-16: - Showcase demos at booth 604 - Talk about #GraphIntelligence for the Databricks Lakehouse with Ryan Knight's presentation. - Attend the Data After Hours at Oracle Park (you can request to attend!) - Have amazing conversations with you all! Is it a date? bit.ly/43hGHIF Check out all the info here 🔼 #DataAISummit

Helping organisations retain control over their data, deployment choices, and future is a vision we share with @Neo4j. For defence and intelligence organisations, data sovereignty is a strategic must. Watch @emileifrem explain why open platforms outperform closed ecosystems.

In this guide, you'll see which tools are useful to build the context you need to avoid hallucinations on your #llms - including orchestration (@LangChain and @llama_index) the memory layer, and tool integration (MCP) Context is key. Start building it with this useful guide. 🚀

ICYMY: We announced Neo4j Virtual Graph: our zero-copy architecture lets you run #Cypher and graph algorithms directly against your data in Snowflake, Databricks, and other cloud data warehouses, without moving a thing. Get graph-powered insights like multi-hop reasoning, and automatically generate a graph model from your existing tables with a single click. 🚨 Virtual Graph is available now in private preview - you can request access by filling in a form bit.ly/4uEeV5h Watch the entire demo here, bit.ly/4dG0AO8 #VirtualGraph #Neo4j

🚀 Mumbai, we’re coming for you! After Open Source Summit India, we’re throwing an intimate 50-person builder dinner, no pitch decks, no panels, just real, unfiltered 2 a.m. debugging-session energy. If you’re building open-source AI products, advocating about security, leading ML teams, or shipping LLMs in production…this room is for you. Speed networking great food the exact people you’ve been wanting to meet. We killed it in Delhi & Bangalore. Mumbai, your turn. June 17th, 5:30 PM onwards. Partners: @neo4j @Qualcomm_Dev Community: @OpenSSF Who’s in? Drop a comment or DM for the link. @SavantNimit @carrycooldude @ramiyengar @alexandererdl

Your Text-to-SQL agent is reading your entire database schema on every single request. Laurent Tande benchmarked a YAML agent against a Neo4j semantic layer agent on the same queries. The result: 20 to 30% fewer tokens on average, up to 10x fewer on simple queries, and roughly 10 points higher accuracy on complex multi-table joins. The difference is how the context is fetched. Instead of dumping the full schema into the prompt, the graph retrieves only the tables, columns, join paths, and business terms relevant to the question. Demo runs in 3 minutes. Code on GitHub. bit.ly/4uX0U2J

Neo4j GenAI Plugin: embed, search, generate x.com/i/broadcasts/1nJOLLzba…

Is your data usable for AI agents? Join @BarrasaDV and learn how to build a semantic layer on Neo4j with ontology at the core. The session is engineering-focused: how to design and evolve ontologies that reduce ambiguity, promote shared understanding across teams, and scale with your systems as they grow. When meaning is explicit, and relationships are first-class elements in your data model, AI-driven applications become more accurate, more maintainable, and easier to trust. Don't miss it! You can register to any of these options: Asia Pacific: bit.ly/4usu48P Europe: bit.ly/4vumOKm Americas: bit.ly/49LKlOy #Ontology #Neo4j #Semantics #AI

Benjamin Squire modeled the SumoDB in #Neo4j and ran Eigenvector Centrality to find the most connected rikishi. The finding: rank does not always mean centrality. bit.ly/49IrvrF #Neo4j #GraphDataScience

Don't get me wrong, I still love code, but sometimes it's great to focus on an agent's tools without fiddling with importing packages, or the agent boilerplate of a specific framework. @neo4j Aura Agent FTW

The 2nd Neo4j Workshop for the Kenya AI Challenge 2026 Hackathon is live at the @BCentreNBO. 30 builders in the room going deeper into Neo4j and graph databases for AI in Agriculture solutions. @KenyaAIHack @MC_AgriFin @neo4j @DirectEdDev #KenyaAIChallenge2026 #AIKenya #Neo4j #AgriTech

about the 3 memory layers for AI agents ⏬ @akollegger ❇️ Short-term memory ❇️Long-term memory ❇️Reasoning memory What are they? Why are they necessary? youtu.be/TdL_W9e9ayU

We've all experienced it: most AI agents forget things because their memory was never designed to last. A conversation buffer stores what was just said. A static knowledge base tells general facts about the world. Neither explains why it made a decision three sessions ago, what it has already tried and found to be wrong, or how a specific entity connects to the task at hand right now. Some time ago, Foundation Capital identified context graphs as a significant architectural trend in the infrastructure of an agentic system, and they solve this with three connected memory layers. 👌 Long-term memory holds enterprise knowledge, the slow-moving facts about your domain that ground the agent in reality. 👌👌Short-term memory captures conversation history, what the user asked, what the agent did, and what it still needs to do, so it does not lose track mid-task. 👌👌👌Reasoning memory stores decision traces, the actual record of why the agent made a choice, not just what it did. The 3 layers are connected so agents can tie together what they know, what was discussed, and what they decided to make better decisions over time, and explain them when they need to. Time to see it in practice: You can build your own context graph with Neo4j Agent Memory, an open-source library that runs on top of any Neo4j instance and packages all three types of memory in the context graph. bit.ly/4x2mZ15 When aiming for enterprise-grade reliability, which deficiency is more costly for your organization: an agent that hallucinates, one that forgets what it’s doing, or both? Answer below (and check out the blog about context graphs!) Build with Neo4j Agent Memory: neo4j.com/labs/agent-memory/ #Neo4j #ContextGraphs #AgentMemory #AgenticAI #KnowledgeGraph

“Have you had your Hume training yet?” That’s the ambition: to be a standard in intelligence analysis worldwide. Together with @neo4j, we’re building tech that helps analysts understand complexity, uncover connections, and protect what matters most. 🔗 hubs.ly/Q04k4S3w0