Resumind

Resumind is an advanced, AI-powered resume builder and analyzer designed to help users craft professional, ATS-friendly resumes. It leverages modern web technologies and powerful LLMs to provide intelligent feedback, content generation, and resume optimization.

🚀 Tech Stack

Frontend (apps/web)

• Framework: Next.js 15 (App Router)
• Language: TypeScript
• Styling: Tailwind CSS, Radix UI, Framer Motion
• State Management: Zustand
• Authentication: NextAuth.js
• PDF Generation: @react-pdf/renderer

Backend (apps/backend)

• Framework: FastAPI
• Language: Python 3.12+
• Package Manager: uv
• Database ORM: SQLModel (SQLAlchemy)
• Migrations: Alembic
• AI/LLM: LangChain, LangGraph, Google GenAI (Gemini), Mistral AI
• Observability: Langfuse
• Background Tasks: Celery with Redis
• Tools: Tavily (Search), Firecrawl (Web Scraping)

Infrastructure

• Database: PostgreSQL 16 (with pgvector for embeddings)
• Caching/Queue: Redis 7
• Containerization: Docker Compose (production) + docker compose watch (local dev)
• Deployment: Dokploy on a self-hosted VPS

---

📂 Project Structure

.
├── apps/
│   ├── backend/        # FastAPI application (AI logic, API, Celery workers)
│   │   ├── src/
│   │   ├── migrations/ # Alembic migrations
│   │   ├── Dockerfile
│   │   ├── Dockerfile.celery
│   │   └── pyproject.toml
│   └── web/            # Next.js frontend (UI, Auth, PDF rendering)
│       ├── app/
│       ├── components/
│       └── Dockerfile  # Multi-stage: dev + production
├── docker-compose.yaml          # Production config (used by Dokploy)
├── docker-compose.override.yml  # Local dev overrides (auto-merged)
└── envrc.example                # Environment variable template

---

🛠️ Getting Started

Prerequisites

• Python 3.12+
• uv (Python package manager) → curl -LsSf https://astral.sh/uv/install.sh | sh
• Node.js 18+ and pnpm 10+ (for running the frontend outside Docker)
• Docker & Docker Compose

1. Clone the Repository

git clone https://github.com/ZakariaBaqasse/resumind.git
cd resumind

2. Environment Setup

cp envrc.example .env
# Fill in the required values in .env

Key variables:

Variable	Description
DB_PASSWORD, DB_USER, DB_NAME	PostgreSQL credentials
JWT_SECRET	Backend auth secret
AUTH_SECRET	NextAuth.js secret
GOOGLE_CLIENT_ID/SECRET	OAuth credentials
GOOGLE_API_KEY	Gemini model access
MISTRAL_API_KEY	Mistral model access
TAVILY_API_KEY	Web search tool
FIRECRAWL_API_KEY	Web scraping tool
LANGFUSE_*	LLM observability

3. Start Local Development

The full stack (Frontend + Backend + Celery + DB + Redis) with live reload:

docker compose watch

• Frontend: http://localhost:3003
• Backend API: http://localhost:8000
• API Docs: http://localhost:8000/docs
• Flower (Celery Monitor): http://localhost:5555

docker compose watch uses docker-compose.override.yml to automatically sync source file changes into running containers — no rebuild required for most edits.

Running Services Outside Docker

Backend:

cd apps/backend
uv sync
uv run uvicorn src.main:app --reload --port 8000

Frontend:

cd apps/web
pnpm install
pnpm dev

---

🗄️ Database Migrations

cd apps/backend

# Create a new migration
uv run alembic revision --autogenerate -m "Description of changes"

# Apply migrations
uv run alembic upgrade head

---

🧪 Running Tests

cd apps/backend
uv run pytest

Test environment variables are loaded automatically from apps/backend/.env.test.

---

🚀 Deployment

This project is deployed via Dokploy pointing at this repository. Dokploy reads docker-compose.yaml directly — docker-compose.override.yml is ignored in production.

Set all required environment variables in Dokploy's UI, then trigger a deploy. To deploy manually on any server:

docker compose up -d --build

CI/CD (GitHub Actions)

• Lint workflow — runs Ruff, mypy, and formatting checks on every push/PR touching apps/backend or apps/web
• Migrations workflow — runs alembic upgrade head automatically on push to main