Jev in production › SQL and data pipelines
Jev judges each candidate triple as a yes/no check before it is written to a governed knowledge graph, as a step in the graph-building pipeline. Graph precision is 0.911 over 500 documents (author).

An ontology-driven knowledge-graph builder: design the ontology, build the graph under it, and serve it — search, API, app, and MCP — from one pipeline.
Powered by Jev — the decision (judge) model inside OntoSmith — and built on Neo4j, the graph store underneath, with native vector search.
The stack has two named pillars: Jev judges what is true enough to write, and Neo4j stores the graph — native graph traversal, a vector index, and full-text search in one store. A build requires both.
For the project's own README, linking back here: