GraphNews

The first (official) Monterey Bay Ontology Meetup ⛵️ Great talks and opportunity to think. Better yet, we went analog. Thank you Joshua Shinavier for a great dry erase board and thoughtful planning, that made for a most excellent meetup. And only 8 minute drive from my house | 44 comments on LinkedIn

From Spreadsheet to Substitution Engine: Building an Agricultural Knowledge Graph in Neo4j, Layer by Layer Build an agricultural knowledge graph one public data source at a time, and watch the …

Every few years our industry discovers a new silver bullet. Today, it's context. Knowledge graphs. Ontologies. Semantic layers. Context engineering. They're all valuable, and necessary. But context alone doesn't solve the fundamental problem of how do we ensure data required by downstream consumers, including AI, is trustworthy? AI doesn't replace the need for trustworthy enterprise data. It amplifies it. In many ways, AI needs exactly what BI, analytics, and operational systems have always needed: • Trusted data • Consistent business meaning • Shared definitions • Authoritative master data • Governance The difference is that AI also needs something more. It needs knowledge. For years we've treated data management and knowledge management as separate disciplines with separate architectures, separate teams, and separate technologies. It's abundantly obvious to me that having data and knowledge in separate organizational silos has reached the end of its useful life. AI demands a more holistic approach that integrates these critical disciplines, and the architectures which support them. 20+ years ago some of us were so myopically focused on BI, that allowed the worlds of enterprise search, document management, and knowledge management to completely split away from data. It's now the time to reunite these disciplines with a more cohesive approach focused on both structured and unstructured data across a cohesive enterprise architecture. There are three key layers to this architecture (see diagram below): 1️⃣ The Semantic Foundation - consolidates all of our repositories of terms and definitions into a single business glossary and enterprise taxonomy. This is the world of data catalogs, and its where basic governance policies are documented. 2️⃣ The Truth Layer (MDM) - establishes trusted representations of customers, products, suppliers, locations, and other core business entities regardless of whether the evidence originates in databases, documents, emails, contracts, PDFs, images, conversations, or other unstructured content. 3️⃣ The Context Layer then connects those trusted entities into a coherent enterprise knowledge model that AI can reason over. The context layer consumes trusted data from the truth layer. The result is shared enterprise semantics built on both truth and context. The future isn't data management. The future isn't knowledge management. The future is an integrated enterprise architecture that finally brings them together. #datamanagement #enterprisearchitecture #informationarchitecture | 28 comments on LinkedIn

PostgreSQL 19 adds SQL/PGQ graph queries. See how its graph features compare with Memgraph on exact hop traversal performance.

It took a while but Postgres finally learned to speak graph. Like MS SQL (in 2017!) and Oracle, that doesn't make our beloved elephant a graph database. PostgreSQL 19 ships SQL/PGQ, the ISO standard for property graph queries. You define a CREATE PROPERTY GRAPH over your existing tables and suddenly your relational schema is queryable with pattern-matching syntax (Cypher-ish) instead of chained JOINs. The graph is a read-only view over the same tables SQL has always queried. It's a solution I suggested to various customers in the past (atop MSSQL) and it safes money and deployments. But the architecture underneath hasn't changed. A property graph in Postgres v19 is still a view over relational storage, executed by a planner built for joins and joins compose differently than traversal. Memgraph benchmarked exactly this: exact N-hop reachability over the Pokec dataset (100K vertices, 1.77M edges), 1 to 5 hops, same data on both engines. Postgres moves from milliseconds into seconds while Memgraph stays under 100ms. At higher hops Postgres hits a 30-second timeout, Memgraph finishes in ~259ms. The gap isn't a tuning problem, it's what happens when you bolt a graph query surface onto a storage engine that was never designed to keep the graph as its primary structure. Each hop re-expands the search through join machinery. SQL/PGQ is still a milestone worth taking seriously though, it legitimizes property graphs as a first-class relational concept and will pull a huge SQL-native audience toward graph thinking. Just don't confuse "my database can express graphs" with "my database was built to traverse." #KnowledgeGraphs #GraphRAG #Postgres #KnowledgeAugmentedAI #GraphDatabases ❧ Postgres LPG: https://lnkd.in/emvFXSFQ ❧ Memgraph's benchmark: https://lnkd.in/ezGGGvYP

Most "LLM wikis" use agents, embeddings, and repeated model calls to organize local notes. I built a deterministic alternative: a pure Python compiler that turns messy markdown into a linked, linted wiki using only the standard library. Along the way, I fixed two real bugs, benchmarked the pipeline on two operating systems, and showed why a compiler is often a better fit than an agent for mechanical text organization.

Calling something "agent observability" can be misleading. In many cases, it's still just execution monitoring. It measures whether the agent completed a task, how long it took, and whether it failed, but it offers little insight into whether the agent actually reasoned correctly within the context of your business. 𝐖𝐡𝐚𝐭 𝐞𝐱𝐞𝐜𝐮𝐭𝐢𝐨𝐧 𝐦𝐨𝐧𝐢𝐭𝐨𝐫𝐢𝐧𝐠 𝐜𝐚𝐭𝐜𝐡𝐞𝐬: 𝐥𝐚𝐭𝐞𝐧𝐜𝐲, 𝐭𝐨𝐤𝐞𝐧 𝐜𝐨𝐮𝐧𝐭, 𝐀𝐏𝐈 𝐬𝐮𝐜𝐜𝐞𝐬𝐬, 𝐫𝐞𝐭𝐫𝐢𝐞𝐬 Useful telemetry, genuinely. None of it checks whether the agent's output was grounded in the right definition of "customer" or the right version of "revenue" for your business. 𝐖𝐡𝐚𝐭 𝐢𝐭 𝐦𝐢𝐬𝐬𝐞𝐬 𝐞𝐧𝐭𝐢𝐫𝐞𝐥𝐲: 𝐰𝐡𝐞𝐭𝐡𝐞𝐫 𝐭𝐡𝐞 𝐫𝐞𝐚𝐬𝐨𝐧𝐢𝐧𝐠 𝐰𝐚𝐬 𝐜𝐨𝐫𝐫𝐞𝐜𝐭 An agent can execute flawlessly against a wrong premise, every single time, and every dashboard you own will show green the whole way through. This is where many AI pilots fail to leap into production. The business never gave the agent a governed definition of its own concepts, so every agent is quietly guessing at meaning it was never handed. At small scale, that's a rounding error. At agentic scale, across thousands of calls a day, that guess compounds into liability. 𝐓𝐡𝐞 𝐝𝐢𝐬𝐭𝐢𝐧𝐜𝐭𝐢𝐨𝐧 𝐭𝐡𝐚𝐭 𝐦𝐚𝐭𝐭𝐞𝐫𝐬 Ontology defines meaning; observability validates that the agent's decisions align with it. But most teams have built neither. They've rather built dashboards for execution and called the problem solved. ➡️ Courts are already converging on this. The emerging principle in AI-related liability cases is "omission is liability"; not having a definition on record is treated the same as having a wrong one. ➡️ The organisations closing the trust gap this quarter aren't running newer models. They fixed ontology and observability underneath the model first, then let the model be whatever model. ➡️ 82% of the transformation value enterprises expect from agentic AI depends on capabilities that don't exist yet, per Gartner. That value sits on infrastructure most companies haven't built. This is the spine of our 𝐐2 𝐒𝐭𝐚𝐭𝐞 𝐨𝐟 𝐃𝐚𝐭𝐚 𝐏𝐫𝐨𝐝𝐮𝐜𝐭𝐬 𝐫𝐞𝐥𝐞𝐚𝐬𝐞. This edition also has a piece on RDF-based semantic layers, and a good one on where semantic models end and semantic layers begin. If you're staring at a pile of pilots wondering why none of them has graduated to production, this one's worth a decent few minutes. 🔖 𝐑𝐞𝐚𝐝 𝐡𝐞𝐫𝐞: https://lnkd.in/dj8GgHTe #Ontology #AIObservability #DataArchitecture

Gartner predicts 50%+ of AI agent systems will use context graphs by 2028. Learn what they are, why they matter, and how enterprises can implement them with metadata-driven platforms like Atlan.

Enterprise software has accumulated forty years of data about business outcomes. It has captured almost none of the reasoning that produced them. That gap is now the center of a growing investment thesis among venture capitalists, anchored by a framework called the context graph.

By Kurt Cagle & Chloe Shannon

A practical guide to creating persistent memory for your AI programming assistant that remains intact across sessions using A. Karpathy's LLM-Wiki pattern

AI context graph infrastructure for agent memory. Learn why your ontology, schema, provenance, temporal state, and permissions should remain yours.

Research shows knowledge graphs and graph embeddings improve the quality and accuracy of Gen AI responses. Here’s how to get started.

TL;DR: The same JSON Schema that validates your data can now declare what that data means. We built a ten-keyword extension vocabulary (x-jsonld-*) that promotes instances to JSON-LD (the JSON serial…

Nvidia's reported $400 million acquisition of Kumo AI is a bet that the next wave of enterprise AI value will come from business data, not just documents and code.

Enterprises are no longer asking whether to adopt GraphRAG.

If you have ever worked with data like “customers who bought products” or “users who follow other users,” or “fraud detection” you know…

🫠 Over the past few months, so many new "engineering" terms have come up that it’s becoming incredibly hard to keep up. IMHO, I think a lot of people are trying to put new terms around everything, and it’s making the space far more confusing than it needs to be. There also seem to be several different definitions for every term. Prompt engineering and context engineering were at least reasonably well understood. But now we have harness engineering, loop engineering, and graph engineering, and different people seem to mean completely different things by them. ⛳ For instance, a large number of people describe harness engineering as designing the harness around an agent. That includes permissions, tools, memory, context compaction, evaluations, observability, and the broader system that controls how the agent operates. ⛳ Others seem to use harness engineering to mean working with an existing harness so that you can get the most out of coding agents, which starts to overlap heavily with loop engineering. ⛳ Loop engineering itself also has several definitions. Some people describe it as simply setting up automated checks and conditions that allow an agent to keep iterating. Others use it more broadly to describe how humans, agents, evaluators, and feedback systems interact across inner loops and outer loops. ⛳ Then there’s graph engineering, which adds another layer of confusion. It seems to refer to explicitly designing the states, transitions, branches, and control flow through which an agent or workflow operates. At the most basic level, I really hope these definitions are roughly correct, but who knows? Everyone seems to have their own interpretation. The more I read about these terms, the more I wonder whether all of this was already part of designing agentic workflows. 🫠 Why are we introducing entirely new terms for ideas that were already part of building agentic systems? Anyway, whipped up a quick diagram for your reference and happy to have you debate it too, no hard feelings, we're all confused!

Creating Agency for Non-Human AI Agents Requires Data on Data: How to Extract, Build, and Deploy this Context. Trinity of Deduction, Productisation, and Activation.