Building an AI-Powered Club Member Portal with AWS
Learn how we built an AI-powered Club Member Portal that enables students to securely sign up, log in, and ask questions using trusted club documents. The portal provides source-based AI answers and a safe fallback when information is unavailable, with a plan for deploying the solution on AWS.
# Building a Club Member Portal with AI-Powered Document Search
## Introduction
As part of the AWS Student Builder Groups hackathon, we built a **Club Member Portal** designed to help new and existing student members quickly find information about their AWS Student Builder Group.
The portal provides secure member authentication and an AI-powered chat experience. Members can ask questions about workshops, AWS account setup, Builder Center, hackathon rules, and other club information. The AI assistant answers using the provided club documents and shows the source of each answer.
## What We Built
Our Club Member Portal includes the following core features:
* Member sign-up using email and password
* Member login
* Forgot-password functionality with email reset
* Protected member chat page
* AI-powered question answering
* Document-based knowledge retrieval
* Source citations for every answer
* Safe fallback when information cannot be found in the documents
The chat is available only to authenticated members.
## How the AI Assistant Works
The portal uses the provided Student Builder Groups starter document pack as its knowledge source.
The documents cover topics such as:
* New member information and club directory
* AWS account setup and billing
* Publishing on AWS Builder Center
* Amazon Bedrock basics
* Hackathon rules
* Workshop information
* AWS Lambda patterns
* Student Builder Groups and community information
The application searches these documents before generating an answer. Each response displays the relevant source document and section so that members can verify the information.
If the requested information is not available in the club documents, the assistant does not guess. Instead, it tells the member that the information could not be found and directs them to the campus AWS Student Builder contact.
This follows the project's requirement to answer only from the club documents and show sources with every response.
## User Flow
The basic user flow is:
**Welcome → Sign Up → Login → Member Chat → Ask a Question → AI Answer + Source**
A member first creates an account or logs in. After authentication, they can access the chat page and ask questions about the Student Builder Group.
## Screenshots
### 1. Welcome / Sign-Up Page
Add a screenshot of your sign-up page here.
**Caption:** Member registration interface for creating an account.

### 2. Login Page
Add a screenshot of your login page here.
**Caption:** Secure member login using email and password.

### 3. Forgot Password
Add a screenshot of your forgot-password page here.
**Caption:** Members can request a password reset through their registered email address.

### 4. AI Chat Page
Add a screenshot of your chat page here.
**Caption:** Authenticated members can ask questions and receive answers based on the club documents.

### 5. AI Answer with Sources
Add a screenshot showing an answer with its source document.
**Caption:** The assistant provides a document-grounded answer with source information.

## AI Safety and Reliability
We designed the chatbot to avoid making unsupported claims.
The assistant:
1. Uses only the provided club documents.
2. Displays sources with answers.
3. Does not invent AWS pricing, limits, or policies.
4. Uses a fallback response when relevant information cannot be found.
5. Does not store passwords or private student information.
These requirements are part of the 70% baseline challenge.
## AWS Deployment Plan
Our prototype can be demonstrated locally. For a production deployment, the system could be moved to AWS.
A possible AWS architecture would use:
* **Amazon Cognito** — member authentication and password-reset workflow
* **Amazon SES** — password-reset email delivery
* **Amazon S3** — storage for club documents
* **AWS Lambda** — serverless backend APIs and document-processing tasks
* **Amazon Bedrock** — AI-powered question answering
* **A search/indexing service** — retrieval of relevant document content
* **AWS hosting services** — deployment of the web application
The hackathon allows teams to run the application locally for the demonstration; live AWS deployment is optional and provides bonus marks. The AWS deployment explanation is required in the pitch and Builder Center article.
## What We Learned
Through this project, we learned how to combine authentication, document-based search, AI-assisted question answering, and source attribution into a practical student-focused application.
We also learned that an AI assistant should not simply generate an answer. It should use trusted information, provide sources, and clearly communicate when the required information is unavailable.
## Conclusion
The Club Member Portal provides a simple way for Student Builder Group members to access important club information through a single platform.
Our 70% implementation focuses on authentication, protected member chat, document-based AI answers, source citations, and safe fallback responses. The project can be demonstrated locally and can be extended to an AWS-based architecture for production deployment.
Enjoyed reading this content? Let the author know!
Your likes, comments, shares, and saves help creators reach more builders.
Loading recommendations
Loading article