M+L GREENMACHINE

The GreenMachine is an ecologically conscious product suggestion chatbot. We've built a Gemini powered frontend and a SQR + Python backend to seamlessly integrate into your company's data sets.

Comment 1

Inspiration 🌟

Our project was inspired by the ability to drive environmentally friendly consumer choices. The aim of the GreenMachine is to push consumers towards products that demonstrate higher standards of safety and environmental consciousness. We have created a suggestion tool that businesses can (theoretically) easily integrate into their web storefronts.

What it does 💡

The GreenMachine is two things: a chatbot and an algorithm. The chatbot is meant to appear and query the user after they select a product from a website. Once they do, the chatbot enters a conversation and determines what the customer values from their product. An example would be ADA compliance - if the selected product is ADA compliant, the chatbot will determine through conversation if the user values ADA compliance and then weigh the value accordingly in the algorithm.

GreenMachine is mean to provide businesses working with high numbers of SKUs and daunting levels of search criteria a simple, friendly way to match customers with greener products.

How we built it 🔧

We worked primarily with the Python language, SQLite, and the Google Gemini API to create this hack. We put together and prompted a chatbot to query the customer to determine priorities (between environmental sustainability, product safety, and product match). The AI analyzes its conversation, determines values (between 0.0 and 1.0) for these categories, which are fed into an algorithm based on our sample product data set. The reasoning behind this choice was that scanning a database of thousands of items was inefficient for the Google Gemini API and would end up wasting unnecessary energy.

Challenges we ran into ⚙️

We're both fairly new to Hackathons, especially in regards to software hacks. Throughout the project, we encountered challenges in using SQLite and integrating Google Gemini into our frontend. Through collaboration and problem-solving skills, we were fortunately able to parse through the many challenges we were faced with.

Accomplishments that we're proud of 🎉

  • Working prototype
  • Successful integration of Google Gemini API
  • Met challenge goals we were faced with

What we learned 📚

This experience pushed us both to explore new areas of programming and technology. Linwood didn't have prior experience working with Google Gemini API, but now, he feels confident integrating it into future products. Likewise, Miles has little experience working with SQL, but now he feels excited to use it in future data science projects.

Most importantly though, this experience has taught us the importance of communication and teamwork. We both played to our strengths, helped the other in times of need, and have ultimately gotten much closer as roommates. :)

What's next for the Green Machine 🚀

We hope to refine the Green Machine, increase the number of potential weighted variables, and improve the user experience. In the future, it would be exciting to see it work with a dynamically changing database and integrate into a web-based experience.

Built With

Share this project:

Updates