Inspiration

Losing personal items is one of the most frustrating and costly everyday problems people face. Whether it’s a phone, wallet, or keys, recovering lost belongings often depends on manual lost-and-found systems that are slow, unreliable, and vulnerable to false claims. LootMax was inspired by the need to make the lost-and-found process faster, smarter, and more secure, while reducing the stress and uncertainty that comes with losing something valuable.

How It Works

LootMax is a privacy-first lost and found platform with two primary roles: Users and Admins.

User Flow

Users can create an account and log in.
They can report an item they have lost or found.
Submissions can include:
- A structured form
- An uploaded image
- Or both
When reporting a lost item, users must provide contact information so they can be reached if a match is found.

Admin Flow

All lost and found submissions are reviewed by an Admin.
Admins can accept or deny submissions based on legitimacy.
Accepted lost items appear on a potential match page, where they are compared against found items.
An AI-generated confidence score is shown to help admins evaluate similarity.
Admins can approve or reject a proposed match.

Preventing False Claims

If a lost item is matched, the user must answer an AI-generated verification question based on item features.
If the answer is correct, the match is confirmed.
If a user submits a found item and it matches a lost one, they receive the contact information of the item’s owner.

At no point is the private inventory exposed to users, ensuring strong privacy guarantees.

How We Built It

LootMax was designed using a three-layer architecture to ensure modularity, security, and scalability.

Frontend (Client Layer)

Built with React
Separate interfaces for Users and Admins
Users can submit lost or found items using forms and image uploads
Admins can review, approve, and match submissions

Backend (Processing Layer)

Built with FastAPI
Serves as the central orchestrator between the frontend, AI, and the database
Integrates Gemini AI to:
- Parse text descriptions and images
- Generate semantic feature tags
- Compare lost and found items
- Produce a match confidence score
- Generate verification questions to prevent false claims

The confidence score is conceptually based on Gemeni AI comparing the diffrent tags associated to the item.

Database (Data Layer)

Built with MongoDB
Stores normalized item data, tags, and inquiry records
Images are not stored; instead, they are converted into feature tags by Gemini AI
This improves privacy and reduces long-term data risk

Each layer communicates strictly through REST APIs, enforcing a clean separation of concerns and aligning with modern best practices.

Challenges We Ran Into

We encountered several technical challenges during development, including:

Establishing reliable communication between the frontend and backend
Managing backend and database integration
Debugging issues where database data was not appearing correctly on the frontend
Designing and implementing a secure verification question system

Overcoming these challenges required rapid learning, careful debugging, and strong team collaboration.

Accomplishments That We’re Proud Of

We are proud of building a system that integrates multiple modern technologies, including React, FastAPI, MongoDB, and Gemini AI. Many of these tools were new to us, yet we successfully combined them into a cohesive and functional product. Designing a privacy-first system that actively prevents false claims was a significant achievement for our team.

What We Learned

Throughout this project, we learned:

How to work with MongoDB, a database we were not previously familiar with
How to integrate AI services into a real application workflow
The importance of strict architectural boundaries between system layers
That AI can be used not only for matching, but also for security and verification

This project significantly expanded our understanding of full-stack development and AI-driven system design.

What’s Next for LootMax

Looking ahead, we aim to:

Further strengthen privacy protections and reduce false claims
Improve and expand the verification system
Enhance the admin dashboard with improved tooling and analytics
Explore deployment in real-world environments such as campuses, events, and public spaces

LootMax has strong potential beyond the hackathon, and we envision it evolving into a secure, scalable lost and found solution.