Vogadu is an AI powered chatbot system designed to provide detailed information about cars using AI and RAG (Retrieval-Augmented Generation) technology.
- 🤖 AI-powered chatbot for car-related queries
- 📚 Access to hundreds of thousands of car-related articles
- 🔍 RAG (Retrieval-Augmented Generation) implementation
- 🤝 Multi-platform support:
- Discord bot
- Messenger bot
- 🌐 Hosted on Firebase
- 📊 Vector database (Qdrant) for efficient article retrieval
- LLM:
- LangChain for LLM request handling and RAG implementation
- RAG for accurate information retrieval and response generation
- Database: Qdrant for vector storage
- Hosting: Firebase
- Data Collection: Custom web scrapers for car-related articles
- Bot Platforms: Discord and Messenger
The project requires several credential files to be set up in the src/credentials
directory. You'll need to add the following JSON files based on the example templates:
discord.json
- Discord bot credentialsfirebase.json
- Firebase service account credentialslangsmith.json
- LangSmith API credentials for LLM tracingmessenger.json
- Facebook Messenger bot credentialsmistral.json
- Mistral API credentialsopenai.json
- OpenAI API credentials and model configurationsqdrant.json
- Qdrant vector database credentials
For each credential file, there is a corresponding example file (e.g., firebase.json.example
) with the required structure. Copy these example files and fill in your credentials:
for file in src/credentials/*.json.example; do cp "$file" "${file%.example}"; done
This project is licensed under the MIT License - see the LICENSE file for details.