ResumeAI - AI-Powered Career Platform

ResumeAI is a premium, full-stack AI career platform designed to help job seekers optimize their resumes, tailor them dynamically via chat, generate cover letters, and track application insights. It utilizes a state-of-the-art Retrieval-Augmented Generation (RAG) pipeline powered by Groq LLMs, LangGraph, FAISS vector stores, and Jina AI embeddings.

---

🌟 Core Features

1. 💬 Stateful Interactive Resume Tailor

• Conversational Dialogue Agent: When users issue broad instructions (e.g. "Help me add a new experience"), the chatbot dynamically asks sequential, structured follow-up questions instead of hallucinating missing data.
• Interactive Workspace: Live dual-panel workspace featuring an Undo/Redo state stack, zoom scaling (50% - 150%), and suggested tailoring triggers.
• ATS PDF Generator: Generates a high-fidelity, two-column ATS-friendly ReportLab PDF layout mimicking the preview template on download.

2. 🔍 RAG-Powered Resume Analyzer

• ATS Quality Audit: Breakdown of match scores, strengths, weaknesses, formatting checks, and impact evaluations.
• Job Matching Matrix: Matches resume text against target Job Descriptions using FAISS vector store indexing and Jina AI embeddings (jina-embeddings-v3).
• Skills Gap Analysis: Displays matching/missing keywords, upskilling paths, and structured course suggestions.

3. 🔐 Enterprise Authentication & OAuth 2.0

• Google OAuth 2.0: One-click Google Login and Registration via @react-oauth/google and Google ID Token verification.
• Secure Token Management: Short-lived JWT Access Tokens combined with HTTP-only, rotated Refresh Tokens for multi-device session security.
• Account Security: Account lockout after repeated failed login attempts, password reset workflow, email verification, and CSRF token protection.

4. ✍️ AI Resume Builder & Cover Letter Writer

• Structured Builder: Form-driven UI converting career history directly to clean formatting.
• Cover Letter Generator: Contextual cover letters matched to target roles and company names with one click.

---

📂 Project Directory Structure

resume_AI/
├── backend/
│   ├── app/
│   │   ├── middleware/      # CSRF and OWASP Security Headers middleware
│   │   ├── routers/         # API Endpoint controllers (auth, builder, resume, dashboard)
│   │   ├── services/        # Business logic (RAG pipeline, PDF utils, Google OAuth, Groq integrations)
│   │   ├── config.py        # Centralized Pydantic application settings
│   │   ├── database.py      # SQLAlchemy setup
│   │   ├── models.py        # SQLite / Postgres Database models
│   │   └── schemas.py       # Pydantic validation schemas
│   ├── tests/               # Pytest integration & unit test suites
│   ├── migrate_auth_upgrade.py # Database schema & OAuth migration script
│   ├── Dockerfile           # Backend container config
│   └── requirements.txt     # Python dependencies
├── frontend/
│   ├── src/
│   │   ├── api/             # Axios API client calls & token interceptors
│   │   ├── components/      # UI components (animations, guards, gauges, layout)
│   │   ├── context/         # React Auth, Theme, and Application contexts
│   │   └── pages/           # Pages (Landing, Dashboard, ResumeTailor, UploadResume, Login, Register)
│   ├── Dockerfile           # Multi-stage production Nginx container
│   ├── vercel.json          # SPA rewrite configs for Vercel
│   └── tailwind.config.js   # Tailored theme configs
├── docker-compose.yml       # Complete stack orchestrator
└── README.md                # Project documentation

---

🗄️ Database Architecture (SQLite / PostgreSQL)

The system uses an SQLite instance (backend/app.db) in development, managed via SQLAlchemy ORM.

Key Entities:

1. users: Core user credentials, OAuth profiles (google_id, auth_provider), roles, email verification status, and lockout tracking.
2. resumes: Uploaded raw text, metadata, and generated ReportLab PDF target paths.
3. analyses: RAG comparison outputs (match_score, strengths, gaps, weaknesses, missing_skills, detailed_review).
4. cover_letters: Generated cover letter document records.
5. refresh_tokens & audit_logs: Multi-device token hashes, revocation tracking, and security audit event logs.

---

⚙️ Local Development Setup

Backend (Python FastAPI)

1. Navigate to the backend directory:

   cd backend

2. Create and activate a virtual environment:

   python -m venv venv
   # Windows:
   .\venv\Scripts\activate
   # Linux/macOS:
   source venv/bin/activate

3. Install dependencies:

   pip install -r requirements.txt

4. Create a .env file in backend/ containing:

   GROQ_API_KEY=your_groq_api_key
   GROQ_MODEL=llama-3.3-70b-versatile
   JINA_API_KEY=your_jina_api_key
   GOOGLE_CLIENT_ID=your_google_client_id
   SECRET_KEY=generate_a_long_random_string
   ALGORITHM=HS256
   DATABASE_URL=sqlite:///./app.db
   CORS_ORIGINS=http://localhost:5173,http://127.0.0.1:5173,http://localhost:3000

5. Apply database schema migrations:

   python migrate_auth_upgrade.py

6. Start the API server:

   uvicorn app.main:app --reload --port 8000

Frontend (React + TypeScript + Vite)

1. Navigate to the frontend directory:

   cd frontend

2. Install npm packages:

   npm install

3. Create a .env file in frontend/ containing:

   VITE_API_URL=http://localhost:8000
   VITE_GOOGLE_CLIENT_ID=your_google_client_id

4. Run the development server:

   npm run dev

---

🧪 Testing Suite

We use Pytest with mock clients to test database and router endpoints in isolation.

To run tests:

cd backend
$env:PYTHONPATH="."
.\venv\Scripts\python -m pytest tests/ -v

---

🐳 Docker Deployment

You can build and deploy the entire multi-container service stack locally using Docker Compose:

# Build and run backend and frontend containers
docker-compose up --build

• Backend API: Available at http://localhost:8000
• Frontend SPA (Nginx): Available at http://localhost:80

---

🌐 Production Cloud Deployment

1. Backend (Render)

1. Import repository into Render as a Web Service or Blueprint (render.yaml).
2. Set Environment Variables:
   • GROQ_API_KEY, JINA_API_KEY, GOOGLE_CLIENT_ID, SECRET_KEY, ENVIRONMENT=production.
   • Set CORS_ORIGINS to your production frontend domain (e.g. https://your-app.vercel.app).
3. Run python migrate_auth_upgrade.py during build command.

2. Frontend (Vercel)

1. Import repository into Vercel and set frontend as Root Directory.
2. Set Environment Variables:
   • VITE_API_URL: Your backend Render URL (e.g., https://resumeai-nh1y.onrender.com)
   • VITE_GOOGLE_CLIENT_ID: Your Google OAuth Client ID
3. Add your Vercel domain to Authorized JavaScript origins in Google Cloud Console.