AWS Builder Center

Builder Center

Building EcoLoop: AI-Powered E-Waste Reverse Logistics for Bharat on AWS

Building EcoLoop: AI-Powered E-Waste Reverse Logistics for Bharat on AWS

India generates 3.2M tonnes of e-waste yearly, 95% handled by informal recyclers who lose margins to middlemen. For Bharat Builds Track 03, I built EcoLoop on AWS. Users snap e-waste photos, Amazon Rekognition classifies the hardware, and the system assigns the nearest local recycler for doorstep pickup at fair scrap rates. Built with React on AWS Amplify, serverless AWS Lambda, DynamoDB, SNS alerts, and AWS SAM CLI. EcoLoop turns discarded electronics into clean impact and fair livelihoods.

TANISH KESHARI

Aspiring SWE
India generates over 3.2 million tonnes of electronic waste every year. More than 95% of this is collected and sorted by informal waste workers (local kabadiwalas). However, two big problems remain:
  1. Urban households do not have a transparent, trustworthy way to dispose of old electronics.
  1. Informal recyclers lose most of their earnings to layers of middlemen.
For the Bharat Builds Tour (WeMakeDevs) under Track 03: Waste and Energy, I built EcoLoop — an AI-powered reverse-logistics platform connecting citizens directly with verified informal recyclers.

How EcoLoop Works

  1. Image Upload & Classification: The user snaps a photo of an old gadget (phone, laptop, monitor).
  1. Amazon Rekognition Computer Vision: The image is analyzed in real time to identify the hardware type and estimate fair scrap valuation while showing carbon diversion metrics.
  1. Automated Dispatch: The system assigns the closest informal recycler for doorstep pickup without middleman fees.
  1. Impact Dashboard: Telemetry tracks total kilograms diverted from landfills and direct payouts to informal workers.

The AWS Tech Stack

  • Amazon Rekognition: Runs computer vision label detection to identify devices in real time without provisioning GPUs.
  • AWS Amplify Hosting: Connects directly to our GitHub repository for continuous integration and automated frontend deployment.
  • AWS SAM CLI (Open Source): Used to scaffold and test serverless architecture locally using infrastructure as code.
  • AWS Lambda & Amazon DynamoDB: Serverless compute and on-demand storage for pickup queues.
  • Amazon SNS: Dispatches instant notifications to informal recyclers upon booking.

What I Learned

Building on AWS serverless services allowed me to focus on solving a real environmental problem rather than managing infrastructure. Connecting Amazon Rekognition via the Boto3 SDK took only a few lines of code while delivering production-grade computer vision accuracy.
here is the deployed link :  https://main.dpc4z740lswvt.amplifyapp.com/ 
# aws-amplify# serverless# sustainability# amazon-rekognition
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