The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
Bank Fraud Detection System — AI-Powered Financial Fraud Detection Platform
The Bank Fraud Detection System is an AI-powered machine learning platform designed to identify potentially fraudulent financial transactions and enhance banking security through intelligent risk analysis. It leverages advanced machine learning algorithms, including XGBoost and CatBoost, along with SMOTE-ENN for handling imbalanced transaction datasets, achieving approximately 98.6% accuracy and 98.9% ROC-AUC in fraud detection.
Built using Python, Flask, Scikit-learn, MySQL, HTML, CSS, and Chart.js, the platform features a secure web application with Role-Based Access Control (RBAC), JWT authentication, and audit logging to ensure controlled access and transaction traceability. It integrates an optimized REST API with approximately 85 ms response latency, enabling fast fraud predictions and risk assessment.
The system combines predictive analytics, secure backend architecture, and interactive data visualizations to support efficient fraud identification, improve financial risk monitoring, and demonstrate the practical application of AI in banking security.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Challenges
View all
Related posts
Hi :) This is a journey through asteroids into the heart of a star. Built with Three.js, React Three Fiber, and custom GLSL shaders. Liquid images, interactive particle typography, and a core that opens under your cursor.
Man... this is cool. 😍
A procedural shaded-relief cartographer's table; export the map as PNG.
I built this project to test A procedural shaded-relief cartographer's table; export the map as PNG. The core engineering challenge was sustaining uncompromised 60 FPS rendering using Canvas2D while handling continuous user interaction without framework overhead.
WHAT I WAS TESTING:
I focused on two core engineering benchmarks:
Real-time topographic contour line extraction using Marching Squares
Mountain ridge carving and hydrographic river basin generation
THE TECHNICAL SIDE:
Combines domain-warped fractional brownian motion (fBm) with Marching Squares contour generation. Renders vintage parchment styled topographic maps with seeded coordinates.
Everything executes in the browser using Canvas2D, Procedural Cartography, Contour Line Marching Squares. No third-party frameworks, no heavy runtime bundles, and zero external tracking.
WHAT I LEARNED:
Contour line marching squares often generated disconnected micro-segments. Stitching adjacent line segments into continuous polyline paths allowed smooth canvas stroke rendering.
Finding the sweet spot between mathematical precision and frame budget took careful profiling, but the result runs at a steady 60 FPS even on modest hardware.
TRY IT YOURSELF:
The live build is deployed on Netlify. Open it in your browser, test the controls, and observe the render response times.
If you are building high-performance 3D, physics, or creative web systems and want to collaborate, reach out here on Contra.
Really nice map detail
CARTELLE: The Society for Speculative Cartography.
"High-Performance Creative Web Exploration"
Drag a procedurally inked globe of places that don't exist. An archive of maps of places that do not exist. The Society for Speculative Cartography charts imaginary worlds with scholarly rigor and unreasonable devotion.
I built this project to test Drag a procedurally inked globe of places that don't exist. An archive of maps of places that do not exist. The Society for Speculative Cartography charts imaginary worlds with scholarly rigor and unreasonable devotion. The core engineering challenge was sustaining uncompromised 60 FPS rendering using Three.js (GLB 3D) while handling continuous user interaction without framework overhead.
WHAT I WAS TESTING:
I focused on two core engineering benchmarks:
Precision implementation of procedural globe under continuous interactive user input
Sustained 60 FPS render performance and zero-dependency browser execution
THE TECHNICAL SIDE:
Engineered directly in the browser with Three.js (GLB 3D), WebGL / GLSL Shaders, Canvas2D, Procedural globe. The application runs without third-party frameworks or build-step dependencies, using hardware-accelerated rendering and custom event loops to achieve zero-latency response.
Everything executes in the browser using Three.js (GLB 3D), WebGL / GLSL Shaders, Canvas2D, Procedural globe. No third-party frameworks, no heavy runtime bundles, and zero external tracking.
WHAT I LEARNED:
Balancing the mathematical rigors of procedural globe with tactile browser interaction required tuning event throttling and memory allocation to eliminate all garbage collection stutters.
Finding the sweet spot between mathematical precision and frame budget took careful profiling, but the result runs at a steady 60 FPS even on modest hardware.
TRY IT YOURSELF:
The live build is deployed on Netlify. Open it in your browser, test the controls, and observe the render response times.
If you are building high-performance 3D, physics, or creative web systems and want to collaborate, reach out here on Contra.
Attractive visual design