team4 – CS144 Final Project

🛠️ Setup Instructions

# Clone the repository
git clone https://github.com/Prof-Rosario-UCLA/team4.git
cd team4

# -------------------------------
# AI Agent (Python)
# -------------------------------
cd ai-service
python3 -m venv venv
source venv/bin/activate   # For Windows: venv\Scripts\activate
pip install -r requirements.txt

# Create .env for ai-service
# (Place this in ai-service/.env)
AZURE_OPENAI_ENDPOINT=your_endpoint_here
AZURE_OPENAI_API_KEY=your_key_here
AZURE_OPENAI_API_VERSION=your_version_here
AZURE_SEARCH_SERVICE_ENDPOINT=your_search_endpoint
AZURE_SEARCH_API_KEY=your_search_key

# -------------------------------
# Server (Node.js/Express)
# -------------------------------
cd ../server
npm install

# Create .env for server
# (Place this in server/.env)
MONGO_URI=your_mongoURI_here
PORT=your_port_here
ACCESS_TOKEN_SECRET=your_secret_here
REFRESH_TOKEN_SECRET=your_secret_here

# -------------------------------
# Client (Vite)
# -------------------------------
cd ../client
npm install

# -------------------------------
# Root
# -------------------------------
cd ..
npm install

Single Container Deployment

This project now ships with a single Docker container that bundles the React client, Node.js API and Python AI service. Build and run the container with Docker Compose:

docker compose up --build

Set REDIS_HOST and REDIS_PORT to the address of your Google Cloud Memorystore instance so no Redis container is required.

The separate Dockerfiles inside client/, server/ and ai-service/ have been removed. The root Dockerfile now builds the entire application. Docker Compose is provided for convenience, but you can also build and run the container directly:

docker build -t oversea-app .
docker run -p 3000:3000 -p 8000:8000 --env-file .env oversea-app