Get honest travel insights powered by Reddit discussions and AI analysis.
TripReality scrapes real Reddit travel discussions, analyzes them with AI, and generates honest travel reports so you can plan trips with confidence.
- Backend: https://tripreality.onrender.com
- Frontend: Deployed via Vercel
- Frontend: React, Tailwind CSS, shadcn/ui
- Backend: FastAPI, Motor (async MongoDB), Uvicorn
- AI/LLM: NVIDIA NIM, Groq
- Data: Reddit API (PRAW)
- Database: MongoDB
cd backend
python -m venv .venv && .venv\Scripts\activate
pip install -r requirements.txt
copy .env.example .env
# Fill in your API keys in .env
uvicorn server:app --reload --port 8001cd frontend
yarn install
# Set REACT_APP_BACKEND_URL=http://localhost:8001 in frontend/.env
yarn startSee .env.example in the backend folder for all required API keys and configuration.
The frontend is also packaged as a standalone Electron desktop app that talks to the Render backend.
cd frontend
npm install
npm run electron:buildThis produces installers in frontend/dist/:
- Windows:
TripReality Setup.exe(NSIS) - macOS:
TripReality.dmg - Linux:
TripReality.AppImage
The app bundles the production frontend build and connects to https://tripreality.onrender.com by default.
Set a GitHub token with repo scope, then build:
set GH_TOKEN=your_github_token
npm run electron:buildelectron-builder will auto-upload the installers to the latest GitHub Release.
cd frontend
npm run electron:devThis starts the React dev server and launches Electron pointing to http://localhost:3000.
- Backend is deployed on Render using
render.yaml - Frontend web app is deployed on Vercel
- Desktop app is distributed via GitHub Releases
- Secrets are configured via Render dashboard env vars (sync: false)
Apache 2.0 — see LICENSE