Verda

Inspiration

Online shopping makes it hard to know if a product is actually sustainable or just good at marketing. Most people don’t have time to dig through materials while shopping. Some don't even know which materials are environment-friendly and which aren't. Thats why I decided to build Verda: to make sustainability simple. Just browse like normal, and Verda gives you a quick sustainability score based on materials, manufacturing, etc.. all with brief explanations to this score. The goal is to help people shop smarter and greener without extra effort.

What it does

Verda analyzes your screen while you're shopping online, giving you "Sustainability Scores" for each product you look at based on many features described on the webpage (e.g. materials used, production of the product, etc..) all on a scale from 1-100 + a detailed explanation on why Verda chose its scores for the product. Then, the big final score is just an average of all the little scores on each section(materials, production, etc..)

How we built it

To let Verda "see"/analyze the user's screen, I used a method called "HTML Parsing" which, in short, extracts the HTML strings(text) from the webpage and send it to OpenAI's ChatGPT 4o-mini model (our model behind the scenes).

Challenges we ran into

The biggest challenge I ran into extracting the right info from each website since each of those major shopping websites(Amazon, Zara, H&M) all have different layouts.