AWS Club Member Portal
What we built
We built a Club Member Portal for our campus AWS Student Builder community.
The idea was simple: instead of students searching through different club documents or asking team members for basic information, they can log in and ask the portal chatbot directly.
How it works
The portal combines three major parts:
UI/UX — Mahesh
We designed a clean and simple interface where students can access the portal and interact with the AI chatbot.
Authentication & Backend — Kartheek
We used Supabase for authentication and application data, allowing members to securely log in and use the portal.
AI & RAG — Sivaram
We built the chatbot workflow using n8n.
The chatbot follows this flow:
User Question → Intent Classification → Document Retrieval → RAG Context → LLM → Response
We created a knowledge base from our AWS Student Builder Group documents and used embeddings with vector search to retrieve relevant information.
RAG-based chatbot
Instead of allowing the AI to answer purely from its general knowledge, we first retrieve relevant information from our club documents.
For example, a student can ask:
> "How do I publish on Builder Center?"
The system searches the knowledge base, retrieves the relevant document content, and generates an answer based on that information.
The chatbot also provides the source document used for the answer.
Fallback system
We also implemented a fallback mechanism.
When the required information cannot be found in our club documents, the chatbot doesn't simply respond with "I don't know."
Instead, it directs the student to the appropriate club contact.
This makes the chatbot more useful for real club interactions.
What we learned
This project gave us hands-on experience with:
RAG and vector search
Embeddings
n8n workflow automation
Supabase authentication
Connecting frontend applications with webhooks
AI chatbot development
GitHub and team collaboration
More importantly, we learned how to divide a real project between team members and integrate our individual work into one working prototype.
Team
Team SKM — RGUKT Ongole
KARTHEEK — Supabase & Authentication
SIVARAM — n8n & RAG
MAHESH — UI/UX
Repository
What's next
We plan to further improve the portal with more knowledge sources, better chat history, an admin interface for updating documents, and AWS-based production deployment.
Loading article