GraphNews

🚀 Knowledge Graphs as Grounded World Models for Neuro-Symbolic AI

GitNexus turns a repo into an AST‑driven knowledge graph directly in the browser: Tree‑sitter parses the code, KuzuDB stores structure and relations, and a local model generates embeddings for semantic search over files, symbols, and call chains. Every step … parsing, graph storage, querying … is client‑side, so your code never leaves your machine while you still get Cypher‑style queries, BM25, impact analysis, and architecture exploration against a real graph, not a bag of vectors. On top of that graph, GitNexus exposes an MCP interface and CLI (gitnexus analyze), so agents like Claude Code or Cursor can request relational context (“all call paths touching this function across services”) instead of streaming arbitrary files into an LLM context window. The result is a zero‑server, living context engine for audits, refactors, blast‑radius checks, and multi‑agent workflows … ideal when you want serious AI‑assisted code intelligence, but your compliance model says “the browser tab is the boundary.” #AI #DevTools #CodeIntelligence #WebAssembly #MCP | 28 comments on LinkedIn

Every LinkedIn post that catches fire re: context graphs and ontologies is missing one glaringly obvious thing.

Cognee raises $7.5M seed led by Pebblebed to build the open-source memory layer for AI agents—backed by founders of OpenAI and Facebook AI Research.

SELECT * FROM database_startups WHERE mosaic_change 0; That query returns a surprisingly small result set as mosaic score changes show a dual-track market for next-gen databases: momentum is concentrating in a handful of companies. Like cloud infrastructure before it, databases are moving from broad experimentation to standardization around a few critical workloads. As that happens, the market becomes far less forgiving. Three things separating leaders from laggards: 1) Leaders are tied to expanding workloads Positive Mosaic movement is concentrated in companies aligned with where spend is actually growing: ↳ real-time analytics in production ↳ AI-adjacent workloads (RAG, GraphRAG, observability, agent workflows) ↳ managed, consumption-based deployment as the default 2) Leaders win workflows, not benchmarks Performance is table stakes. The real separation comes from owning: ↳ developer onboarding and time-to-production ↳ day-2 operations and observability ↳ enterprise decisioning and mission-critical systems 3) Commercial execution is the moat Momentum correlates with repeatable enterprise sales, security and compliance at scale, and partner-driven distribution. Strong technology without scalable commercialization shows up directly as negative Mosaic drift. Where the leaders are finding success: → ClickHouse (+24) Momentum is driven by rapid funding velocity, enterprise customer expansion, aggressive hiring, and platform expansion into observability and LLM monitoring tied to AI production use cases. → Supabase (+85) Large Mosaic gains reflect explosive developer growth (4M+ users), rising AI workload adoption, and early enterprise traction as Supabase expands beyond developer tooling into production infrastructure. → Starburst (+35) Gains are anchored in durable enterprise signals: deep penetration in regulated industries, strong partner leverage, and renewed demand for federated analytics amid data-residency constraints. → Neo4j (+17) Momentum reflects scale and defensibility: $200M ARR, 84% Fortune 100 penetration, and accelerating GraphRAG adoption, contributing to one of the highest IPO probabilities we track. → Cockroach Labs (+4) While not a breakout mover, stable Mosaic performance signals resilience via enterprise renewals, IBM OEM distribution, and positioning around AI-era uptime and global scale requirements. Bottom line: momentum belongs to companies winning workflows and distribution, not feature comparisons. Over the next 6–12 months, leaders become magnets for platform partnerships and M&A, while others are pushed into defensive consolidation or niche competition. P.S. Comment "SELECT * FROM cb_insights;" for free access to CB Insights’ predictive intelligence on the companies in next-gen database markets. | 79 comments on LinkedIn

By Oleksii Tkachuk, Kartik Sathyanarayanan, Rajiv Shringi

Authors: Luis Medina and Ajit Koti

Authors: Adrian Taruc and James Dalton

The latest Gartner report on #GraphRAG is out, and the verdict is clear: if you are building AI applications, combining Knowledge Graphs with LLMs is the smartest move you can make right now.

🔮 By 2030, more than one in 10 enterprises will be AI-first, having outperformed competitors through the adoption of agents, semantics and converged D&A platforms.

Film Club Knowledge Graph. Contribute to SteveHedden/fckg development by creating an account on GitHub.

Scientific Reports - Construction and application of knowledge graph for seed quality standard documents

Why I Stopped Letting LLMs Build My Knowledge Graphs (And What I Did Instead) . How Fixed Entity Architecture turned a struggling Graph RAG system into the backbone of an enterprise code migration platform.

This post links to all the 20 posts I created on LinkedIn.

Everyone is racing to solve the same problem: how do you give AI agents the context they need without dumping everything into the context…

Enterprise Data Architecture for the Agentic Era, Part 1 of 3

The Memory Graph Layer for Agents

Finnish schools send their middle school students to companies for a week-long internship to learn about working life. I did my internship…

Joe Reis created a phenomenal visualization for his recently published 2026 Practical Data Community State of Data Engineering report. It included links to the raw CSV data and SQL queries associated with various charts. That simple act — something the Linked Open Data community ethos has always been about — enabled me to build on that work by adding a Knowledge Graph, Ontology, SPARQL, and SPASQL (SPARQL inside SQL)–based rendition using a live Virtuoso database instance for high-performance access, connectivity, and reuse. See comments for links. #CDO #CDIO #CAIO #CIO #CTO #CxO #CMO #SemanticLayer #ContextGraph #KnowledgeGraphs #SemanticWeb #LinkedData #DataVisualization #SPARQL #SQL #SPASQL #VirtuosoRDBMS #HowTo #CSV #PracticalDataSurvey

Agent control is a graph composition problem. Each layer answers a necessary control question. None is sufficient alone. Composing across them is hard.

Connecting the layers of Semantic Architecture

What would it take to finally connect data and meaning?

What "semantics" actually is and what we should focus on

Solution-level siloes and object-level descriptions aren't cutting it

An Introduction to Semantic Model-Driven AI

Knwler — turn documents into a structured knowledge graph, in minutes. We built Knwler to solve a problem every compliance team, legal department, and research analyst knows too well: extracting structured insight from dense, complex documents takes a lot of manual work.

Geospatial data analysis systems are currently very relevant. Most such systems use either relational databases or graph databases. This paper presents the idea of using both approaches, taking into account the main features and advantages of each. A concrete example of a city transport network is used to experimentally examine the use of this hybrid approach. A special ETL procedure was developed to transform data from the corresponding graph database to a relational one, as well as the reverse process from the relational to the graph database. The results show which type of queries are better suited for relational databases, and which for graph databases. Additionally, for certain specific queries and applications, neither database type is capable of providing any results. Although this kind of hybrid architecture has issues with data duplication, the performance gains achieved are significant, making this approach highly efficient.

I am an AI realist, not a skeptic.

Bridging Semantics and Syntax with Annotation-Driven Generation