By: Oscar Sharaz Spencer, Andy Zhou, Nate Wu, Max Zabarka & Mark Havkin
In our world most large companies do not focus on sustainability. Unsustainable companies like Coca Cola, Nestle, and Starbucks produce billions of tons of waste per year, ultimately destroying the fabric of our world. The best way to leverage technology to solve this problem is educating the root supporters of these multi-national monopolies: consumers, and supporting smaller sustainable alternatives.
Before buying a product either online or in person, a user can scan it with Stomata. They receive a full sustainability analysis on it, reporting three sub-category scores and one overall sustainability score. Other users’ reviews and recommendations on the product will be present, as well as a list of more sustainable alternative products with higher scores, in aims of supporting sustainable companies.
Our team developed a custom multimodal model that excels in product detection, leveraging the ESG dataset. This collaborative effort utilised both JavaScript and Python to build and refine the model's capabilities.
Our team encountered some tough challenges in integrating many components, not just AI, but across the entire tech stack. The complexities of aligning various systems and ensuring seamless integration across layers of the stack presented some big hurdles, though we persevered and learned along the way, resulting in a completed application.
We designed the frontend and backend, integrating AI and datasets to measure sustainability scores within Stomata, proof of our team's technical skill and commitment to sustainability through innovation.
Through Stomata, we learned that change can be achieved through technological applications, and that with careful planning and sophisticated architecture, it's possible to integrate complex technical systems effectively.
Our aim is to support businesses which factor in the well-being and sustainability of the planet, so we hope to expand our app and improve user experience.* This includes expanding the datasets* used and improving our AI software. If the app is eventually pushed to consumers, our biggest goal is to make it easy and comfortable to scan any item the consumer wants.
esg-dataset flask gpt-4 javascript material-ui python react.js scss Try it out