Main Features

• Agentic AI Resume Generation Pipeline

  At the heart of ResumeX lies an agent-based queue system powered by BullMQ. The resume generation workflow is broken into specialized agents such as Job Description Analyzer, Skills Ranker, Project Ranker, and Description Generator. Each agent executes its task independently in a background worker, passing the result to the next, ensuring modularity, scalability, and high throughput.

• Smart Resume Personalization with Gemini

  ResumeX uses advanced LLM capabilities via Gemini to tailor resume content based on the provided job description and the user's professional data. The system generates role-specific summaries, optimized skill matching, and project descriptions that closely align with employer expectations—all in real-time.

• Scalable and Modular Backend Architecture

  The backend is designed using NestJS with a clean separation of concerns across modules, services, and repositories. MongoDB acts as the primary datastore for user profiles and resumes, while Redis powers real-time job queues and heartbeat monitoring. This architecture ensures ResumeX can handle high job volumes and scale horizontally without performance degradation.

• Centralized Quota Management and Rate Limiting

  A dedicated Gemini Layer, implemented using the Singleton Pattern, centrally manages API usage. It ensures compliance with global and user-level usage limits, handles retry logic for failed AI requests, and protects third-party resources from overuse, maintaining system stability under load.

• Secure and Robust API System

  ResumeX emphasizes security with JWT-based authentication, role-based access control, and custom NestJS guards to protect sensitive routes. DTOs and validation pipes are used to enforce strong typing and data integrity, while encrypted storage and best practices ensure user information remains protected throughout the system.