https://github.com/devsrijit/orpheus-chat
https://github.com/devsrijit/orpheus-chat
Last synced: 9 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/devsrijit/orpheus-chat
- Owner: DevSrijit
- Created: 2024-07-29T09:49:03.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2025-03-01T01:00:55.000Z (over 1 year ago)
- Last Synced: 2025-03-06T15:50:25.054Z (over 1 year ago)
- Language: Python
- Size: 3.96 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Orpheus Chat: Behind the Scenes
Orpheus Chat is an AI-powered Slack bot that brings Hack Club's knowledge base to life. Let me walk you through how I built it. You can start talking with Orpheus by pinging `@Orpheus AI` at Slack.
## The Journey
I wanted to create a bot that could truly understand and respond to questions about Hack Club's diverse community. I started by researching existing bots and AI models, but I quickly realized that none of them were tailored to my specific needs.
So, I set out to build my own.
### Building the Knowledge Base
First, I needed to give the bot deep knowledge about Hack Club. I wrote a Python scraper that collected data from the HackClub/dns repository to find all domains, then I scraped all the sites, gathering information about various projects, initiatives, and community resources. This raw data was then transformed into a structured JSON format (`hackclub_finetune.jsonl`), making it perfect for AI consumption.
### Making the Bot Smart
To give the bot real understanding, I used advanced AI techniques:
- Generated semantic embeddings from the processed data
- Stored these embeddings in Pinecone's vector database
- Integrated Pinecone's assistant feature for intelligent responses
### Real-time Communication
The bot comes alive through Slack integration:
- Built using the Slack Bolt Framework for robust event handling
- Implemented Socket Mode for secure, real-time communication
- Added comprehensive error handling and logging
- Hosted on nest
### Technical Stack
- Python for the core implementation
- Slack Bolt for the messaging interface
- Pinecone for AI capabilities and vector storage
- MongoDB for structured data management
The result? A smart, context-aware assistant that helps Hack Club community members get the information they need, when they need it. This bot is hosted on nest ! Feel free to ping `@Orpheus AI` to start a chat or ask a question.