Getting Started
Deploy SentiNext on your own infrastructure and start analyzing Steam reviews.
Deployment: Self-Hosted
SentiNext runs on your infrastructure. Choose Docker Compose for quick setup or manual deployment for full control.
Docker Compose (Recommended)
One-command deployment with PostgreSQL, backend, and frontend.
# Clone and configure
git clone https://github.com/NicoCampa/SentiNext.git
cd SentiNext
cp .env.example .env.local
# Edit .env.local with your LLM API key
# Start everything
docker compose up --build
This starts PostgreSQL, the FastAPI backend, and the Next.js frontend. The dashboard will be available at http://localhost:3000.
Manual Setup (Alternative)
Run the frontend and backend separately for development or custom deployments.
# Backend
conda env create -f environment.yml
conda activate SentiNext
./run_backend_local.sh
# Frontend (separate terminal)
./run_frontend_local.sh
See LOCAL_DEVELOPMENT.md in the repository for full setup instructions.
Analyze Your First Game
Search a game and start a review analysis run.
- > Search for a game by name or paste a Steam App ID.
- > Start the analysis: the dashboard uses the latest 1,000 reviews by default.
- > Wait for completion: you'll see progress, then a dashboard with insights and quotes.
- > Explore & share: use Reviews, Chat, Compare, Reports, and exports when needed.