Einscribe project

HEC Digital Entrepreneurship certificate

Find any artist behind an unidentified artwork on the internet.

If you have any questions that are beyond the scope of this help file, Please feel free to email me: fherbine@student.42.fr.

What's Einscribe ?

During a month we had to build our own mocked company at HEC Paris in teams. First we had to come up with a valid problem statement, ours was:

"It's very time consuming for recruiters to find credits for digital works."

Einscribe is a browser plugin that reveals the creator behind any digital work and enables recruiters to hire.

This website allows you to try a minimal version of Einscribe that we realised during our last week at HEC.

Installation

This plugin has been tested on Mozilla Firefox (101.0)

1. Download the WebExtension here
2. Unzip einscribe.zip and extract all its content on your computer.
3. Open Firefox, and type about:debugging as a URL.
4. Click on This Firefox > Load Temporary Add-on...
5. A popup will open (file explorer), search for manifest.json in the unzipped folder.
6. The Einscribe icon should be in the upper right corner of your browser, each time, you want to authenticate an artwork click on it.

Status of this project

How it works ?

To collect data, Einscribe relies on a Backend, on the other end, the front-end enables user to query this backend.

Technical overview

Einscribe needs both a backend that gets reliable data all over the internet and a frontend that enables the end user to match his image to our database.

Technical architecture

In order to collect these reliable data, we had to developped a strong back-end architecture.

1. Data is collected by our web crawler named "CrawlerStein" from well known designers' platforms.
2. CrawlerStein struggles to find additional data about the artists (their website, browser, third-party tools, ...)
3. This data is stored into our database.
4. This data can be reinforce, by either scripts (AI models or not) or human like us.
5. Finally the database can be queried through our API : The "ScribeProvider".
6. Actually, the Mozilla web extension will use this API.

How many data collected ?

For the purpose of this project we have only collected 14k images & 1500 designers, yet CrawlerStein is currently able to retrieve 200 images / minute.

Want more ?