CVCraft

AI-Powered ATS Resume Builder

Build professional, ATS-optimized resumes with AI-powered content refinement, real-time scoring, and instant PDF export — completely free, no signup required.

---

✨ Features

Feature	Description
🤖 AI Content Refinement	Improve resume sections using NVIDIA Gemma-2 AI
📊 Real-time ATS Scoring	Instant ATS compatibility scores with matched/missing keyword analysis
📄 Professional PDF Export	Download ATS-optimized PDFs in 2 proven formats
👁️ Live Preview	See your resume update in real-time as you type
🎨 2 ATS Templates	Jake Ryan Professional & Clean ATS format
🆓 No Signup Required	Completely free, no account needed
📈 GA4 Analytics	Built-in Google Analytics 4 tracking
🔍 SEO Optimized	Open Graph, Twitter Cards, JSON-LD structured data

---

🛠️ Tech Stack

Frontend

Backend

AI Models (NVIDIA NIM)

Model	Purpose
gemma-2-2b-it	Resume content refinement & generation
nv-embed-v1	Semantic text embeddings for ATS scoring
rerank-qa-mistral-4b	Relevance reranking for keyword matching

---

🚀 Getting Started

Prerequisites

• Node.js 18+ & Yarn
• Python 3.10+
• MongoDB (local or Atlas)
• Redis (optional — app works without it locally)

Backend Setup

cd backend
python -m venv .venv
.venv\Scripts\activate        # Windows
# source .venv/bin/activate   # macOS/Linux
pip install -r requirements.txt

Create backend/.env:

MONGO_URL=mongodb://localhost:27017
DB_NAME=cvcraft
CORS_ORIGINS=*

# NVIDIA API Keys (get from build.nvidia.com)
GEMMA_API_KEY=nvapi-xxxxx
EMBED_API_KEY=nvapi-xxxxx
RERANK_API_KEY=nvapi-xxxxx

# Local dev: relaxed rate limits
ANON_RATE_PER_MIN=100
ANON_DAILY_LIMIT=1000

Start the server:

uvicorn server:app --host 0.0.0.0 --port 8000 --reload

Frontend Setup

cd frontend
yarn install

Create frontend/.env:

REACT_APP_BACKEND_URL=http://127.0.0.1:8000

Start the dev server:

yarn start

Open http://localhost:3000 🎉

---

📡 API Endpoints

Method	Endpoint	Description
POST	/api/ai/refine	AI-powered text refinement
POST	/api/ai/generate	General AI text generation
POST	/api/ai/latex	Generate LaTeX from CV data
POST	/api/ats/score	Enhanced ATS scoring with term breakdown
POST	/api/pdf/compile	Server-side PDF compilation (requires pdflatex)
POST	/api/ai/ats	Legacy ATS scoring
GET	/api/	Health check

---

🧪 Testing

Backend (42 tests — Hypothesis property-based + pytest)

cd backend
.venv\Scripts\python -m pytest tests/ -v

Frontend (42 tests — fast-check property-based + Jest)

cd frontend
npx craco test --watchAll=false

84 total tests covering:

• ATS score bounds & fallback behavior
• Cache deduplication
• Backward compatibility
• AI toggle behavior
• PDF compilation
• Integration flows

---

📁 Project Structure

CVCraft/
├── backend/
│   ├── server.py            # FastAPI app entry point
│   ├── ai_router.py         # AI endpoints (refine, generate, embed, rerank)
│   ├── ats_router.py        # Enhanced ATS scoring endpoint
│   ├── pdf_router.py        # PDF compilation endpoint
│   ├── template_engine.py   # LaTeX template rendering (6 templates)
│   ├── auth_router.py       # Authentication (unused — app is free)
│   ├── limits.py            # Rate limiting & caching utilities
│   ├── redis_client.py      # Redis connection (with in-memory fallback)
│   ├── email_sender.py      # Email utilities
│   ├── requirements.txt     # Python dependencies
│   └── tests/               # Property-based & unit tests (42 tests)
├── frontend/
│   ├── src/
│   │   ├── App.js           # Main app with routing & GA4
│   │   ├── pages/
│   │   │   ├── Home.jsx     # Landing page with SEO
│   │   │   └── CreateCV.jsx # CV editor with AI features
│   │   ├── components/
│   │   │   ├── CVPreview.jsx        # Live CV preview (2 templates)
│   │   │   ├── AIToggle.jsx         # AI refinement toggle
│   │   │   ├── TemplateSelector.jsx # Template picker
│   │   │   ├── ATSScorePanel.jsx    # ATS score display
│   │   │   └── SEOHead.jsx          # SEO meta tags
│   │   ├── lib/
│   │   │   ├── api.js        # Backend API client
│   │   │   └── analytics.js  # GA4 tracking utilities
│   │   └── __tests__/        # Frontend property tests
│   ├── public/
│   └── package.json
└── README.md

---

📸 Screenshots

Home Page	CV Editor

| Landing with features & CTA | Real-time editor with AI + ATS |

---

🔑 Environment Variables

Variable	Required	Description
MONGO_URL	✅	MongoDB connection string
DB_NAME	✅	Database name
GEMMA_API_KEY	⚡	NVIDIA Gemma-2 API key (AI features)
EMBED_API_KEY	⚡	NVIDIA nv-embed-v1 key (semantic ATS)
RERANK_API_KEY	⚡	NVIDIA rerank key (semantic ATS)
REDIS_URL	❌	Redis URL (optional — in-memory fallback)
REACT_APP_GA4_ID	❌	Google Analytics 4 measurement ID
CORS_ORIGINS	❌	Allowed CORS origins (default: *)

⚡ = Required for full AI features. App works without them (graceful fallback).

---

📄 License

MIT © Kartik Sharma

---

Built with ❤️ by Kartik Sharma