MailSorterAI

Inspiration

Our inspiration came from a personal problem. Every time I opened my Gmail (which, honestly, was only about once a month), I was overwhelmed by the amount of newsletters, promotions, and marketing emails filling my inbox. Important messages like missing assignments, account deactivations, subscription renewals, and other time-sensitive emails were buried under all the clutter. We wanted to create a solution that helps people focus on what actually matters.

What it does

MailSorterAI uses the Gmail API to access your emails, analyzes them using a Groq AI model, and automatically organizes them into meaningful categories. Instead of manually sorting through hundreds of emails, users can quickly find the messages that matter most.

How we built it

Essa Khan focused primarily on backend development, while Sam Gupta and Bryson Walker worked on the frontend. For the backend, Essa used MongoDB because its document-based structure closely matches the way email data is organized. The frontend was built using React, JavaScript, and CSS to create an interactive and user-friendly experience.

Challenges we ran into

One of the biggest challenges we faced was being completely unfamiliar with the Gmail API and Google Cloud. Learning how authentication, permissions, and API access worked took significant time.

Another challenge was choosing an AI model. Initially, we planned to use the Gemini API because of its power and availability through Google AI Studio. However, we ran into API limitations and payment requirements. Since none of us had a payment method available, we switched to Groq's free AI models, which allowed us to continue development.

Accomplishments that we're proud of

One major accomplishment we're proud of is seeing the final product actually work. Watching MailSorterAI successfully connect to Gmail and sort emails was incredibly rewarding, especially considering we only had about one hour left before finishing.

What we learned

Throughout this project, we learned a lot about the Gmail API, Google Cloud, authentication, and working with AI models. Essa also gained experience with Vite and React while helping understand how the frontend and backend connect together.

What's next for MailSorterAI

In the future, we want to improve MailSorterAI by making the AI model faster, more accurate, and better at understanding different types of emails. We may even explore training our own model using a custom dataset. The possibilities are endless.

Built With

• api
• html
• javascript
• native
• node.js
• react
• style.css
• tailwindcss
• vite