We're looking for organisations or individuals who would be interested in piloting LumosAI / IxGraph — and would welcome anyone who wants to explore what's possible with a causal semantic graph layer across enterprise data.
What we've built
IxGraph is a semantic graph database built on Neo4j, structured around an 11-domain ontology with 544 observable conditions. It connects data from HR, engagement surveys, finance, and operational systems into a single causal intelligence layer — without any source transformation or schema editing. The raw source data stays as-is; the ontology does the interpretive work.
What makes it interesting (from a Neo4j perspective)
The graph naturally surfaces cross-domain relationships that are invisible in siloed systems — not just correlations but causal chains. We're using Neo4j's native API layer as the query backbone, and have built two interface layers on top of it:
- A standard BI-style dashboard — D3 force-directed graph visualisation, causal explorer with three analysis modes (Causal Chain, Impact Analysis, Path Trace)
- A natural language chat interface powered by the Anthropic SDK (Claude), connected directly to the Neo4j API — so users can query the graph in plain English and get explainable, structured responses
What we're looking for in a pilot partner
Honestly, anyone curious. Ideally an organisation that:
- Has data sitting in multiple disconnected systems and wants to see what a unified graph layer reveals
- Is interested in explainable AI over organisational or operational data
- Wants to explore what a Neo4j-native semantic ontology looks like in practice
Equally, if you're an individual working on something adjacent — causal graphs, HR analytics, enterprise knowledge graphs — and want to dig into the architecture, we're open to that too.
Happy to share
The ontology structure, the IxGraph data model, and the dashboard / API interface design are all things we can walk through. Drop a reply here or DM — always good to talk to people building on Neo4j.