rkhoja/chatbot-rag-alliance repository overview

⁠🤖 Helpy – Alliance Cluster Concierge Chatbot

A containerized, Retrieval-Augmented Generation (RAG) chatbot built to support users on Canada's national HPC systems. Powered by Flask, LangChain, and vector search (RAGFlow), with multi-model support (OpenAI, Ollama, Claude, GroqCloud, Gemini).

Maintained by:
Rahim Khoja⁠ ([email protected]⁠)
For U of A Research Computing and Alliance cluster support

⁠📦 Quick Start

docker pull rkhoja/chatbot-rag-alliance:latest

⁠🧰 Features

• Flask + SocketIO WebSocket API
• Markdown + Streaming LLM responses
• LangChain with support for OpenAI, Groq, Anthropic, Ollama, Google AI
• Real-time document search via RAGFlow (or Ragie)
• Prompt templating, persona switching, session state
• Ready for Kubernetes or standalone Docker

⁠🚀 Deployment

Helpy can be deployed via:

• Kubernetes with TLS (Traefik or NGINX Ingress)
• Docker Compose or raw Docker (for dev/testing)

See: KUBERNETES-DEPLOYMENT.md⁠

⁠🔄 CI/CD

This image is automatically built and pushed via GitHub Actions from:
https://github.com/ualberta-rcg/chatbot-rag-alliance⁠

Tags reflect the source branch used (e.g., latest).

⁠🧠 About Us

Helpy is maintained by the University of Alberta Research Computing Group⁠, supporting Digital Research Alliance of Canada clusters and users.

⁠📜 License

MIT License — free to use, modify, and redistribute.
View License⁠