Jev Neo4j

A Python research project that uses Jev (TypeSafe AI, System 1) with Neo4j to traverse cybersecurity knowledge graphs probabilistically, choosing promising graph edges and scoring nodes against a goal instead of generating text.

Kervin Hu's Neo4j Developer Blog experiment runs Jev Noul over 1,822 blocker-filtered candidate pairs from 703 synthetic contact records: F1 0.943 at a 0.5 threshold, about two cents per thousand pairs, and a hybrid reaching 0.972.

A demo application that navigates a Neo4j graph one hop at a time: Jev gets the outgoing relationships as a Choice plus a goal Noul in one system_one call, and beam search follows the probabilities.