Inspiration:

The idea for GreenInbox came from noticing how much of our daily carbon impact is hidden in digital receipts, from every flight, meal delivery, and package shipment. Yet few people manually track this data because it’s scattered across dozens of emails. We wanted to build something that uses AI for good (in this case, environmental awareness), and something that works automatically without manual input.

What it does:

GreenInbox automatically connects to a user’s Gmail account, finds emails that likely contain purchase or travel information, and estimates the carbon footprint of each activity.

How we built it:

We used the following stack and services:

  • Gmail API → to fetch emails containing keywords like “receipt,” “order,” “flight”
  • Google Sheets API → to log results and visualize the data
  • Node.js → for writing the backend scripts
  • Google Cloud Console → for API authentication and OAuth setup

Challenges we ran into:

We faced a few key challenges during development:

  • Authentication hurdles while connecting to Gmail and Sheets APIs for the first time
  • Categorization accuracy: designing simple keyword logic that correctly detects activities
  • Time pressure: making sure the demo was working and visually clear in under 36 hours
  • Team coordination: balancing learning time for APIs with coding progress Despite these, we successfully delivered a working demo that impressed judges with its clarity and real-world value.

Accomplishments that we're proud of:

Our team is very proud of many milestones we achieved during the hackathon:

  • Functional end-to-end demo: From Gmail to Google Sheets, the full automation worked live during our presentation.
  • Real-world impact: The project tackles climate awareness using everyday data that people already have: their email receipts.
  • Team collaboration: Every team member contributed, from API integration and classification logic to data visualization and presentation design.
  • Beginner-friendly build: None of us were experts in Google APIs before, but we learned to set up authentication, parse data, and use cloud services efficiently.

What we learned:

This project helped us learn:

  • How to use Google APIs securely and effectively
  • How to divide a complex system into beginner-friendly roles
  • How sustainability tech can be made accessible and automated We also learned key principles of data parsing, API authentication, and project documentation.

What's next for GreenInbox:

We plan to expand GreenInbox by:

  • Sending a weekly carbon summary email to users
  • Allowing users to adjust emission factors based on lifestyle
  • Deploying the script via Google Cloud Run for automation

Built With

  • chatgpt
  • gmail
  • gmailapi
  • googleadk
  • googlecloudconsole
  • googlesheets
  • googlesheetsapi
  • lookerstudio
  • python
  • vertex
  • vscode
Share this project:

Updates