Open Source · MIT Licensed

Fraud Intelligence Platform

Transparent. Explainable. Deterministic.

An open-source ecosystem of fraud intelligence, investigation tooling, device risk, regulatory mapping, and explainable decision engines — designed to work together as a cohesive platform.

9
Public repositories
6
Interactive demos
2
Decision engines
3
Investigation tools
MIT
Open-source license

Why this platform exists

Context before the details.

Most fraud repositories demonstrate isolated concepts. This platform brings together decisioning, investigations, analytics, device intelligence, regulatory mapping, and operational controls into a cohesive ecosystem of explainable fraud intelligence projects. Every component is designed to be transparent, auditable, and human-review first.

Currently Building

ZIC Decision Engine ATO Investigation Standard Android Device Risk SDK Emergency Outflow Lock AI Agent Risk Evaluation Suite

Platform Modules

Grouped by purpose so the ecosystem is easy to navigate.

Featured Case Study

A synthetic mule network investigated end-to-end on the platform.

Investigating a Synthetic Mule Network

1,247 transactions · 12 entities · 3 fraud rings · Overall risk 95/100 · Model confidence 87%

Multi-ring detection with entity relationship graph, AI recommendation on Ring #2, risk factor breakdown, and investigator workspace for accept / override decisions.

Live Demos

Interactive previews of the core modules. Click through to explore.

Fraud Investigation Canvas

Interactive analyst workspace for multi-entity fraud investigations with explainable AI recommendations.

Fraud Analytics Dashboard

Operational monitoring UI for fraud losses, detection rates, alert queues, and case investigation workflows.

Emergency Outflow Lock

Decision layer that defines when to lock outflows, what to restrict, and why — with full audit lifecycle.

Android Device Risk SDK

Silent device risk intelligence at KYC and disbursal — nine fraud signals, sub-200ms, zero PII on device.

Regulator Intelligence Library

Multi-jurisdiction regulatory change tracking with priority scoring, rollout phases, and client applicability.

ATO Investigation Standard

Structured account takeover methodology reconstructing full attack chains from signals to monetisation.

Architecture

How the modules connect as a single ecosystem.

Signals Deterministic Decision Engine Investigation Canvas Analytics Dashboard Regulator Library Emergency Outflow Lock Device Risk SDK AI Agent Risk Suite

Core Principles

The philosophy behind every module.

Transparent decision-making
Deterministic risk scoring
Explainable outputs
Human-review first
Audit-ready decisions
Synthetic demonstration data
Open standards
Privacy-conscious design

Built With

Technologies used across the platform.

Python FastAPI JavaScript HTML5 Chart.js Kotlin REST APIs GitHub Pages GitHub Actions Explainable AI Fraud Detection Risk Scoring Graph Analysis Mermaid

Roadmap

What is shipped and what is next.

Completed

  • Explainable Decision Engine
  • Investigation Canvas
  • Analytics Dashboard
  • Device Risk SDK
  • Emergency Outflow Lock
  • Regulatory Intelligence
  • ATO Investigation Standard
  • AI Agent Risk Suite

Planned

  • Case Replay Engine
  • Network Intelligence
  • Graph Database Integration
  • Live API Gateway
  • SaaS Portal

Open Source

Independent research in fraud intelligence.

9
Repositories
MIT
Licensed
6
Live demos
Aug 2026
Last updated

Built by

The person behind the platform.

Gururaj G J

Gururaj G J

Fraud Risk · Merchant Risk · Financial Crime · Trust & Safety

Fraud Intelligence Specialist and Risk Systems Researcher based in Bengaluru. Building open-source tools for explainable decisioning, investigations, and device risk.

Interested in

Fraud Engineering Trust & Safety AML Merchant Risk Risk Consulting Financial Crime

Let's connect.