AWS Builder Center

Memora - Your Second Memory

Memora is an AI-powered personal memory engine that transforms scattered documents into searchable facts, connected information, and proactive reminders, helping people find and remember what matters.

CSE Student
Memora: Building a Personal AI Memory Engine
The Problem: We Have Too Much Information to Remember
Have you ever searched through hundreds of screenshots, PDFs, emails, and documents just to find one important piece of information?
You remember that you saved your laptop invoice, warranty card, college fee receipt, or an important certificate. But you don't remember where it is or what exactly it says.
Our digital lives are filled with valuable information, but finding and remembering that information is still a frustrating, manual process.
What if we could build a second memory for ourselves?
That's the idea behind Memora, an AI-powered personal memory engine that turns scattered documents into searchable, structured, and useful memories.
Our Solution
Memora allows users to upload personal documents such as:
- Invoices and receipts - Warranty cards - Certificates - Internship and offer letters - Travel tickets - Insurance documents - Important screenshots
Instead of simply storing these files, Memora uses AI to understand their contents and extract useful information.
For example, when a user uploads a laptop invoice, Memora can identify:
- Product name - Purchase date - Purchase amount - Seller - Warranty information - Important tags
This information is stored as structured memory, making it easier to retrieve later.
How Memora Works
Our application follows a simple workflow:
1. Upload
The user uploads a PDF or image through the website.
2. Understand
An AI model analyzes the document and extracts relevant information.
3. Store
The extracted facts, summary, and document details are stored in a database.
4. Search
The user can ask natural-language questions about their uploaded information.
5. Remember
Memora identifies important dates and displays reminders for information the user may forget.
Example Use Case
Imagine uploading a laptop invoice.
Instead of searching through your downloads folder months later, you can ask:
«"How much did my laptop cost?"»
Or:
«"When does my laptop warranty expire?"»
Memora retrieves the relevant information and displays the source document so users can verify the answer.
The goal is not just to create another document chatbot.
The goal is to make personal information more accessible, understandable, and actionable.
Technology Stack
We are building Memora using a modern web stack designed for rapid prototyping and cloud deployment.
- Next.js: Frontend and backend application - Tailwind CSS: User interface styling - Gemini API: Document understanding and question answering - Supabase: Database and file storage - AWS Amplify: Application deployment and hosting - GitHub: Version control
We chose this approach to focus on building a functional product quickly while deploying the application on AWS.
Our architecture can evolve to use additional AWS services as the project develops.
Our Key Features
1. AI-Powered Document Understanding
Convert unstructured documents into useful structured information.
2. Personal Memory Search
Ask questions about uploaded files using natural language.
3. Connected Information
Find related documents and information about the same product, event, or topic.
4. Important Date Discovery
Identify deadlines, renewals, and expiry dates that users may want to remember.
5. Source-Based Answers
Show the relevant source document alongside answers to help users verify information.
What Makes Memora Different?
Document chat applications already exist, and we recognize that personal knowledge management is not a new concept.
Our focus is on combining document understanding with structured memory and proactive discovery.
Instead of asking users to remember filenames or search through folders, Memora aims to help them discover information they have forgotten and surface important details at the right time.
We are starting with a focused prototype and plan to improve its intelligence, reliability, and usability through testing.
Our AWS Deployment
AWS plays an important role in making our application accessible online.
We are deploying our web application using AWS Amplify Hosting.
As the project evolves, we can integrate additional AWS services for document storage, processing, and AI workflows.
Our goal is to build a practical application that demonstrates how cloud technologies and AI can solve everyday problems.
Future Scope
We envision Memora becoming more than a document search tool.
Potential future improvements include:
- Automatic document categorization - Smarter semantic search - Personalized reminders - Document relationship mapping - Secure integrations with other information sources - Improved support for multilingual documents - Privacy-focused personal information management
These features would be developed gradually, based on user feedback and practical requirements.
Conclusion
Our digital lives are full of information that matters, but remembering where everything is can be difficult.
With Memora, we are exploring how AI can transform scattered documents into a personal memory system that helps users find, understand, and remember important information.
Memora — Your second memory.
We are building this project as part of the WeMakeDevs x AWS hackathon, focusing on practical problem-solving, rapid development, and cloud deployment.
Any opinions in this article are those of the individual author and may not reflect the opinions of AWS.
Enjoyed reading this content? Let the author know!

Your likes, comments, shares, and saves help creators reach more builders.

Loading recommendations

Loading article