BellBoost Al Companion

By CODE Atzen at START Hack 2024 in St. Gallen

#1. Place in Intel Liftoff Challenge @ START Hack 2024

Introduction

An App that uses AI to speak to the Bell Food Group employees and motivates them to do their jobs. The app is available in every language and made for people without technical knowledge. Employees can ask questions about the company, get life and health advice, learn languages, and find coworkers and groups of the same interests. The app uses AI to match people and make their lives in a new country more accessible and fun.

Features

• Chatbot
• Multi-language
• Chat with coworkers
• Match people
• Learn languages
• Health advice
• Life advice
• Company information

Technologies

Backend

• Python
• Poetry
• FastAPI
• Predictionguard LLM
• Chatengine IO API

Frontend

• React-Native
• Expo

Sample Video (Dev Stage Mar 22. 2024)

https://github.com/Hannes221/bellfood_start_2024/assets/113210660/276287c4-3b45-48e8-9c26-2d5b7f78ca76 https://github.com/Hannes221/bellfood_start_2024/assets/113210660/4db1f2da-a8dd-44cf-ac49-bef3e389c473

Installation

The app is containerized with Docker. To run the app, you must have Docker installed on your machine.

1. Clone the repository
2. Copy the .env.example file in the frontend and backend directories to .env cp .env.example .env
   • Fill in the credentials for Predictionguard LLM and Chatengine IO Api in the .env file of the backend directory
   • Fill in the Chatengine IO Project ID in the .env file of the frontend directory
3. Run npx expo in the frontend directory
4. Potentially run yarn install in the frontend directory
5. Run poetry install in the backend directory
6. Or simply run docker-compose up in the root directory

Backend Endpoints

• /message - Chat with the AI via a POST request in llm.py
• /authenticate - Chatengine IO access via POST request in chat.py
• /user - Coworker # via POST request in interest.py
• /matching - Coworker Matchmaking via POST request in interest.py

Predictionguard LLM

The AI is provided by Predictionguard LLM.

Ask @Hannes221 for the credentials.

Chatengine IO API

The Chat feature is provided by Chatengine IO. The project address is here.

Ask @maxonary for the credentials.