Helpful

Mobile and Web Development for Plastic Container Recycling Startup to Make a Difference for the Whole World

Сlient

Helpful is a London startup designed to get people motivated to switch to environmentally recyclable bottles and reject using single-use bottles. This application is aimed at rewarding users for leaving plastic bottles and coffee cups for recycling in specially designated locations. The project has a potential to attract investments in the cleaning of aquatic areas and beaches.

The concept of the company is to apply artificial intelligence, machine learning, and digital currency technologies that would change the world for the better.
AIS Novations Project Manager

Challenge

AIS was entrusted with the task to develop applications for Android and iOS, and launch a website with embedded Azure Custom Vision Service.

Azure Custom Vision is the perfect tool that allows building, deploying, and improving image identifiers. Such identifiers apply labels representing classes or objects to images according to their visual characteristics.
AIS Novations Dev Team
Team
2Developers
1Project Manager
Nov 2017 — Nov 2019

Solutions

  • The project has been realized based on Apache Cordova for two platforms: Android and iOS.
  • Azure cloud platform has been used to make the platform scalable.
  • The «live search» of the object in the frame is implemented through TensorFlow models.
API did not suit for “live search” implementation due to their insufficient speed. That is why we have used TensorFlow models. They allow for faster launching and run on any device.
AIS Novations Dev Team

Framework
ASP.NETAngularJS
Miscellaneous
Microsoft AzureApache CordovaIonic
When working with Cordova, we faced compatibility issues, lack of required support and plugin functionality. We have edited the plugins written in the stacks that were native to each platform.
AIS Novations Dev Team

Results

The technology developed allows for the implementation of the entire startup idea: recognize the logo of the container manufacturer and display the material type and recycling methods; identify the user location and lay out the route to the nearest recycling location; accumulate Helpful-tokens and exchange them for eco goods.

The app recognizes containers of more than 20 brands: Coca-Cola, Pepsi, Sprite, Tropicana, Starbucks, etc.

Client’s feedback

Other projects