Nicolas Agostini NicolasAgostini
← Back to Projects

ResumeSmith

A proprietary AI-powered resume analysis tool co-founded by Nicolas Agostini via Orobyte IT.

Overview

ResumeSmith is an AI-powered resume analysis platform designed to help users tailor application materials for target positions. Built with secure authentication, Nginx reverse proxy, Stripe integration, and Gemini AI. As a co-founder and lead backend engineer, I played a key role in its development, focusing on backend architecture, frontend development, and database management.

Key Features:

  • Resume analysis against job postings using Gemini AI.
  • AI-generated interview questions based on resume and job description.
  • Secure Authentication: Email/Password login and Google SSO with automatic profile picture handling.
  • Multi-language support (English, Portuguese, Spanish).
  • Subscription management with Stripe (Free tier with limited analyses, Premium tier with additional features).
  • Export options: PDF format with proper styling and Word document format.

Technologies Used

  • AI: Gemini AI
  • Frontend: React
  • Backend: Python (Flask), Gunicorn
  • Database: PostgreSQL
  • Infrastructure: Nginx, Docker Compose, Let's Encrypt (HTTPS)
  • Payments: Stripe

The system ensures secure data handling and a seamless user experience, from initial sign-up to exporting a polished resume.

System Architecture

The ResumeSmith platform is designed with a microservices-oriented architecture, containerized using Docker and orchestrated with Docker Compose. This setup ensures scalability, maintainability, and efficient resource management. Below is a simplified diagram illustrating the main components and their interactions:

ResumeSmith System Architecture Diagram - Nicolas Agostini GCP Cloud Architect

Key components include a React frontend, a Python (Flask) backend API for core logic and AI integration, a PostgreSQL database for persistent storage, and Nginx as a reverse proxy and for serving static assets. Stripe integration handles payment processing.