TruNews

README

This document outlines the technical details of TruNews, a news outlet project designed for the "Software Engineering 2" class.

Mission: Delivering unbiased news with integrity and a seamless user experience, free from intrusive ads.

Features:

• User-friendly interface:
  • Read articles
  • Share articles on social media and with QR codes
  • Save articles
  • Follow people and communities
  • Participate in events
• Advanced AI:
  • Automatically generate natural titles and categorize articles (models not included in this repo)
• Personalized content:
  • Discover trending articles and related communities
  • Search for specific users, articles, and communities
• Community-driven:
  • Join communities centered around diverse interests
  • Join events that center around those interests

Technical Stack:

• Programming Language: TypeScript
• Database: PostgreSQL
• API Communication: REST
• Notable Algorithms:
  • Feed: Softmax based on categories
  • Trending Articles/Authors: Weighted algorithm with sigmoid functions
  • Related communities: Statistical deviation algorithm
  • Searches and related articles
• Queue System: RabbitMQ

Infrastructure

• Fully dockerized
• Unit tested with Jest
• GitHub Actions for automated server management
• Logging with Winston
• Prisma ORM
• Entry validation with ZOD
• Database population scripts
• Trending algorithm reset scripts
• Swagger documentation
• Image conversion with Sharp
• Dependency injection with Tsyringe
• Deployment:
• DockerHub hosting (all code and docker-compose)

AWS infrastructure

• EC2 server
• S3 for images
• CloudFront for CDN
• Lambda and EventBridge for simulating user activity and trend generation
• Nginx proxy for SSL certificate

Contact Information:

David Alfonso Cañas | Backend Software Developer inquiries: david.alfonso.canas@gmail.com