Facts Detector-0
ASSET_VIEWER
1 / 6
Full Stack

Facts Detector_

Facts Detector Bangladesh is an AI-driven misinformation detection platform built to identify and mitigate fake news across digital content. The system enables real-time verification of text, images, and videos using machine learning models and automated analysis pipelines. It is designed with a scalable full-stack architecture to handle high-volume requests efficiently, with a strong focus on performance, security, and reliability. Targeted primarily at the Bangladesh digital ecosystem, the platform helps improve information integrity on social media and news channels while promoting digital literacy through accessible fact-checking tools.

01 / Stack Architecture
TypeScript
Nextjs
NestJS
MongoDB
Redis
CI/CD
Docker
AWS
02 / Core Specs
  • [01]AI-powered fake news and misinformation detection
  • [02]Text, image, and video verification system
  • [03]Real-time analysis with confidence scoring
  • [04]Object detection using YOLO-based models
  • [05]User-friendly verification dashboard
  • [06]REST API integration with scalable backend
  • [07]Optimized performance using Redis caching
  • [08]Secure and scalable architecture with NestJS
  • [09]Cloud deployment using AWS & Docker
  • [10]CI/CD pipeline for automated deployment
  • [11]SEO optimized and high-performance Next.js frontend
  • [12]Fully responsive and cross-device compatibility
03 / Role
  • Designed and developed full-stack architecture (frontend, backend, and database)
  • Built AI-powered verification workflows for text, image, and video analysis
  • Developed scalable REST APIs using NestJS
  • Integrated machine learning inference pipeline for misinformation detection
  • Implemented Redis caching for performance optimization
  • Designed and built admin dashboard for system monitoring and control
  • Handled authentication, authorization, and secure API access
  • Managed deployment using Docker and AWS infrastructure
  • Optimized frontend performance and SEO using Next.js
  • Ensured system scalability and production-level stability