FraudShield
Real-Time Enterprise Banking Fraud Detection Platform
End-to-end real-time enterprise banking fraud detection platform with a 99.93%-accuracy ML model, live transaction streaming, 3-tier RBAC (Analyst/Investigator/Admin), device fingerprinting, anomaly tracking, and case management.
System ID
02
Fraud-Detection-Banking-Dashboard
Model Accuracy
M-0199.93%
Keras ML Classifier
Access Control
M-023-Tier RBAC
Analyst/Inv/Admin
Core Services
M-033 Runtimes
Next.js/Spring/FastAPI
Telemetry
M-04Live Stream
Supabase Realtime/RLS
Section 01
Problem Statement
Banking fraud operations require low-friction coordination between live transaction ingestion, ML risk scoring, device fingerprinting, and role-scoped case investigation.
Section 02
System Constraints
- >Separate high-throughput transaction rules/RBAC from Python ML scoring
- >Enforce strict 3-tier row-level access control across Analyst, Investigator, and Admin roles
- >Stream live anomalous transactions and case updates to investigators
Section 03
System Architecture & Dataflow
Transaction & Device Stream
Java 21 Spring Boot 4
FastAPI + Keras (99.93%)
Next.js 16 + Supabase RLS
Section 04
Key Engineering Decisions
- 01Paired a Java 21 / Spring Boot 4 enterprise backend with a FastAPI + Keras fraud scoring service achieving 99.93% accuracy.
- 02Enforced 3-tier RBAC (Analyst, Investigator, Admin) backed by Supabase Realtime and Row-Level Security.
- 03Combined device fingerprinting, anomaly tracking, and case management inside a Next.js 16 + React 19 dashboard.