---

📸 Screenshots

---

✨ Features

• 🤖 Local & Remote AI (Ollama + Fallback Support): Works out of the box with local or remote Ollama models (qwen3.5:9b, llama3, gemma4, etc.) with full privacy, plus optional server-side fallback support.
• 📄 Resume Ingestion & Auto-Parsing: Upload existing PDF, DOCX, or TXT resumes and automatically extract structured personal info, work experience, skills, education, and projects.
• 🗂️ Master Baseline Context: Maintain an exhaustive, single source-of-truth career profile with all your roles, achievements, and projects.
• ✨ Section-Level AI Refinement: Edit any experience or section manually or iterate with targeted AI prompts ("Rewrite bullets with STAR metrics", "Emphasize distributed systems architecture").
• 🎯 Job Tracker & Smart Tailoring:
  • Track applications across a status pipeline (Saved, Applied, Interviewing, Offer, Rejected).
  • Paste a job description to compute candidate match scores and keyword alignment.
  • AI intelligently selects, reorders, and tailors relevant bullets to fit a clean, targeted resume without inventing unsupported skills or metrics.
• 📐 ATS-Friendly LaTeX & PDF Export: Generates clean, battle-tested LaTeX source code and instant PDF previews for both Resumes and Cover Letters.
• 🧠 PostgreSQL + pgvector: Unified database storing both relational records and semantic vector embeddings.

---

🚀 Quickstart

Prerequisites

• Node.js (v20+) & npm
• Docker & Docker Compose (runs PostgreSQL pgvector and the backend API container automatically)
• Ollama (optional, for local AI inference)

Step-by-Step Setup

1. Clone the repository and cd into it

git clone https://github.com/naveedkhan1998/resume-forge.git
cd resume-forge

2. Install dependencies first

npm i

3. Configure environment variables (optional)

Default development values work out of the box, or you can copy .env.example to customize ports, database credentials, or LLM endpoints:

cp .env.example .env

4. Start the development servers

npm run dev

That's it! npm run dev uses Nx to concurrently launch:

• Frontend Web App: http://localhost:4200 (React + Vite with hot reload)
• Backend API & Database: http://localhost:8000/api/docs (Dockerized Django Ninja API + PostgreSQL pgvector with automatic database migrations on startup)

---

Alternative: Running the Backend Directly on Host (Without API Docker Container)

If you prefer running the Python backend directly on your host machine using uv:

# 1. Start only the PostgreSQL (pgvector) database container
docker compose up -d db

# 2. Create a Python 3.12 virtual environment and install backend dependencies
uv venv --python 3.12 .venv
uv pip install -e apps/api

# 3. Run database migrations
npm run migrate

# 4. Run frontend and local backend
npm run dev:web        # Terminal 1: http://localhost:4200
npm run dev:api:local  # Terminal 2: http://localhost:8000

---

📜 Available Scripts

All commands are run from the project root after running npm i:

---

🏗️ Architecture & Monorepo Layout

Managed with Nx Monorepo:

resume-forge/
├── apps/
│   ├── web/                    # React 19 + Vite + Tailwind CSS v4 + Lucide UI
│   └── api/                    # Django 5 + Django Ninja + PostgreSQL (pgvector)
├── docker/
│   ├── Dockerfile.backend      # Python 3.12 + uv + Gunicorn production/dev image
│   └── entrypoint.sh           # Auto-enables pgvector extension & runs migrations
├── docs/
│   └── screenshots/            # Application UI screenshots for documentation
├── .github/
│   ├── ISSUE_TEMPLATE/         # Bug report & feature request templates
│   ├── PULL_REQUEST_TEMPLATE.md
│   └── workflows/              # CI/CD workflows (Docker Hub automated builds)
├── docker-compose.yml          # PostgreSQL (pgvector) + Backend API services
├── package.json                # Monorepo scripts & dependencies
├── CONTRIBUTING.md             # Contributor guide & workflow
├── CODE_OF_CONDUCT.md          # Contributor Covenant Code of Conduct
├── SECURITY.md                 # Security policy & vulnerability reporting
└── LICENSE                     # MIT License

---

🤖 Ollama Configuration

1. Ensure your Ollama service is running and pull a model:

   ollama run qwen3.5:9b
   # Or any model of your choice:
   # ollama run llama3:latest
   # ollama run qwen2.5-coder:14b
   

2. Inside the ResumeForge web UI, open AI Settings to test connectivity and select your preferred model.
   • Host Native API: Use http://localhost:11434
   • Dockerized API (npm run dev): Use http://host.docker.internal:11434 (configured automatically in docker-compose.yml)
3. Recommended Settings: For lightweight models, use a temperature between 0.1 and 0.3. ResumeForge requests structured JSON, validates required fields, and automatically performs a focused repair pass if a model returns malformed JSON.

---

🌐 Deployment

Backend Docker Image

A prebuilt backend image is published via GitHub Actions to Docker Hub:

docker pull naveedkhan1998/resume-forge-backend:latest

Frontend Cloudflare Pages Deployment

The frontend (apps/web) can be deployed directly to Cloudflare Pages:

# Preview production build locally with Wrangler
npm run pages:preview

# Deploy to Cloudflare Pages
npm run pages:deploy

Cloudflare Dashboard Git Integration Settings:

• Framework preset: None / Vite
• Build command: npm run build:web
• Build output directory: dist/apps/web
• Root directory: /

---

🤝 Contributing

We welcome contributions from the community! Whether it's fixing bugs, adding resume templates, improving AI parsing prompts, or enhancing documentation:

1. Read our Contributing Guide for development setup and PR conventions.
2. Please follow our Code of Conduct.
3. For security issues, please refer to our Security Policy.

---

📜 License

Distributed under the MIT License. See LICENSE for full details.