ApplyAI — AI Job Application Assistant

ApplyAI is a cloud-native, AI-powered assistant that helps streamline the job search. The MVP focuses on shortening application time, increasing application volume, and improving the quality of each submission by providing AI-driven resume tailoring and conversational career guidance.

Table of contents

• Overview & Goals
• Architecture & Tech Stack
• Core features
• Data model (Firestore)
• Getting started (local development)
• Deployment notes
• Next steps
• License

Overview & Goals

ApplyAI helps job seekers automate and improve their job applications. The MVP aims to:

• Reduce application time by automating tailored materials.
• Increase application throughput by making it easy to generate high-quality applications.
• Maximize job-fit relevance by emphasizing the most relevant skills in tailored resumes.

Architecture & Tech Stack

This project is a full-stack, decoupled application.

[User Browser (localhost)] <--> [Next.js (React) Frontend] <--> [Firebase Auth]
          |
          v
[FastAPI Backend (Local/Cloud)] <--> [Gemini API]
          |
          v
[Firestore Database]

Stack & Rationale

• Frontend: React (Next.js) + TypeScript + Tailwind CSS
  • Why: A modern, production-grade React framework for fast, type-safe, and beautifully styled component-based UI.
• Auth: Firebase Authentication (Google OAuth)
  • Why: Secure, easy-to-implement identity management that integrates natively with Firestore.
• Backend: Python + FastAPI
  • Why: High-performance, asynchronous-first framework that's perfect for I/O-bound tasks like calling external AI APIs.
• AI Model: Google Gemini API
  • Why: High-quality generation for conversational chat and complex text-rewriting tasks.
• Database: Google Firestore (NoSQL)
  • Why: A serverless, scalable document database for storing unstructured data like chat history.
• DevOps: Docker + Google Cloud Run
  • Why: A containerized, serverless-first deployment that scales to zero.

Core features

1. Authentication (Frontend Implemented)

   • Secure Google Sign-In via Firebase.
   • Global session management using React Context.

2. Resume Tailoring (Full-Stack Implemented)

   • POST /resumes endpoint that accepts a base resume and job description.
   • Generates a tailored resume using Gemini and renders it as formatted Markdown.
   • Saves the result to Firestore linked to the user.

3. AI Chat Agent (Backend Implemented)

   • POST /chat endpoint that takes a user message and returns a Gemini-powered response.

Data model (Firestore)

• chats (collection)

  • chatId (document)
    • userId (string)
    • messages (array of { role: 'user'|'ai', content: string })

• tailored_resumes (collection)

  • resumeId (document)
    • userId (string)
    • originalResume (string)
    • jobDescription (string)
    • tailoredResume (string - Markdown)
    • createdAt (timestamp)

Getting started (local development)

This repository is a monorepo containing both the client and server.

Prerequisites

• Python 3.12+
• Node.js 18+ and npm
• A Google Cloud project with Firestore, Gemini API, and Firebase Auth enabled.

1) Backend (server/)

1. Navigate to the server directory:

   cd server

2. Create and activate a virtual environment:

   python3 -m venv .venv
   source .venv/bin/activate

3. Install dependencies:

   pip install -r requirements.txt

4. Configure environment variables: Create a .env file in server/ with:

   GEMINI_API_KEY="your-gemini-api-key"
   GOOGLE_APPLICATION_CREDENTIALS="path/to/service-account.json"
   

5. Run the backend locally: (Run from root folder)

   PYTHONPATH=$PYTHONPATH:$(pwd)/server uvicorn server.main:app --reload --port 8000

   API docs: http://127.0.0.1:8000/docs

2) Frontend (client/)

1. Navigate to the client directory:

   cd client

2. Install dependencies:

   npm install

3. Configure environment variables: Create a .env.local file in client/ with:

   # Point to your local Python backend
   NEXT_PUBLIC_API_URL=[http://127.0.0.1:8000](http://127.0.0.1:8000)
   
   # Firebase Configuration (Get these from Firebase Console)
   NEXT_PUBLIC_FIREBASE_API_KEY=...
   NEXT_PUBLIC_FIREBASE_AUTH_DOMAIN=...
   NEXT_PUBLIC_FIREBASE_PROJECT_ID=...
   NEXT_PUBLIC_FIREBASE_STORAGE_BUCKET=...
   NEXT_PUBLIC_FIREBASE_MESSAGING_SENDER_ID=...
   NEXT_PUBLIC_FIREBASE_APP_ID=...
   NEXT_PUBLIC_FIREBASE_MEASUREMENT_ID=...

4. Run the frontend locally:

   npm run dev

   App: http://localhost:3000

Deployment notes

The backend is deployed to Google Cloud Run. The frontend is currently local but can be deployed to Vercel or Firebase Hosting.

• CORS: The FastAPI app allows requests from http://localhost:3000. For production, update server/main.py with the deployed frontend domain.

Next steps

• Render the tailored resume as Markdown.
• Implement full user authentication (Firebase Auth).
• Connect Frontend User ID to Backend (Save real user data).
• Implement the frontend UI for the /chat endpoint.
• Add a full CI/CD pipeline with GitHub Actions.

License

See the LICENSE file for details.