RAVEL System Uses TigerGraph and AI to Detect Complex Payment Fraud Rings in Milliseconds

Two developers, Nikhil Kumar Panigrahi and Sai Manohari Godavarty, built RAVEL — an autonomous fraud investigation platform — for the TigerGraph Hacker House Goa challenge. The system combines TigerGraph Cloud's compiled graph queries, LangGraph-based AI agents, and entropy-driven evidence evaluation to identify coordinated payment fraud rings. Traditional fraud tools and tabular models like XGBoost can miss such schemes; RAVEL's multi-hop graph traversal uncovered a 35-card device collusion ring behind a seemingly routine $125 transaction. The platform completed graph traversals in 0.238 seconds compared to 24.2 seconds for standard methods, achieved zero hallucinations, and recovered all 20 fraud ring cases with full policy compliance. It also automates regulatory report drafting with a dual-key human approval workflow, cutting analyst investigation time significantly.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)
Log in to join the discussion and vote.
Log in