StartUp Crew

Using CrewAI to validate startup ideas from different points of view.

Ollama

This project includes an Ollama stack for running local LLMs and a web UI for interaction.

Setup

1. Copy the environment template:

   cp ollama/.env.template ollama/.env

   Edit ollama/.env to adjust ports, model name, or other settings as needed.

2. Start the stack:

   cd ollama
   ./run.sh

   This script will:

   • Load environment variables from .env
   • Start the required Docker Compose services
   • Wait for Ollama to be ready
   • Pull the specified model (from $OLLAMA_MODEL in .env)
   • Tail the logs for all services
   • Automatically stop and clean up services when you exit the script (Ctrl+C)

Services

• Ollama API: http://localhost:11434 (or the port you set in .env)
• Web UI: http://localhost:3000 (or the port you set in .env)

Stopping

• Press Ctrl+C in the terminal running run.sh to stop all services and clean up.