SpeedyBots

An educational chatbot designed for German-speaking train drivers at Schweizerische Südostbahn AG (SOB). This chatbot assists drivers in learning and communicating in technical Italian, which is essential for railway operations in the Ticino region.

Requirements

Backend:

• Python 3.8 or higher
• Flask
• Ollama (running in http://localhost:11434)

Frontend:

• Node.js 10.0 or higher
• npm or yarn

Database:

• SQLite

How run the project

Backend

Navigate to the backend directory:

cd backend

Create a virtual environment:

python -m venv venv

Activate the virtual environment: •Windows:

venv\Scripts\activate

•Mac/Linux:

source venv/bin/activate

Install the dependencies:

pip install -r requirements.txt

Start the Flask server:

python run.py

Open http://localhost:5000 to view it in your browser.

DB

!!! Optionally, there is also a demodb (app.db) in the repository.

Navigate to the backend directory:

cd backend

Initialize the database

flask db init

Apply existing migrations to create the database schema:

flask db upgrade

Frontend

Install the dependecies:

npm install

Start the frontend:

npm start

Open http://localhost:3000 to view it in your browser.

Next steps

• Improve the reliability of the LLM
• LLM integration to vary exercise feedback responses