BioMatrix ( Senior Category 10 - 12 )

Team Logo
Main Pic
Side Pic
Kashyap(Left) and Arav(Right) reviewing the final website
Anshuman(Left) and Atharv(Right) fixing a bug

Inspiration

Climate change is an overwhelming issue for most of us, but taking individual action is often complicated. When we built this, we wanted to build something that makes sustainability, simple, personal and actionable. Our inspiration was the idea that if people could get assistance and immediate feedback on their daily habits from their room setup to their waste management, they would naturally make better choices, leading to an actionable and aware audience.

What it does

BioMatrix is an AI-powered sustainability assistant that helps anyone reduce their damage to the environment with four connected modules:

-GreenRoom: Uploading a room photo is all you need to do to get a sustainability score and personalized tips. It is an AI Room Analyzer. -CarboPrint: It calculates your carbon footprint based on your lifestyle, diet, commute, and daily habits. -GreenSort: An AI Waste Classifier, using computer vision to classify waste into recyclable, compostable, bio-degradable, non-biodegradable, etc. with disposal tips. -GreenGaurdian: A Sustainability tracker that tracks all your actions and provides you with a score as well as visual analysis.

How we built it

-For GreenRoom, we used a Gemini API key with a model 2.5 pro, and used it for two components of our project: Component 1 was CV and the other was Generative AI. It uses camera based input to identify the key aspects of the room and suggest changes to protect the environment. -For CarboPrint, we used React and forms to help the users fill out a questionnaire to help derive their carbon footprint through an algorithm. -For GreenSort, we used another key to take input of the user via camera based input. It uses flask for frontend. -For GreenGuardian, we used python and streamlit to host a webapp where users can go through a form and then this runs it through an algorithm and provides suggestions for sustainable construction practices. -All projects are hosted either on flask or streamlit.

Challenges we ran into

-Standardizing carbon emissions across different lifestyles and habits was difficult because datasets vary between sources. -Building a clean, unified UI for four different tools within the time limit. -Ensuring the Waste Classifier worked with real-life, messy images rather than perfect dataset samples.

Accomplishments that we're proud of

-Building four fully functional AI-powered modules in under 24 hours. -Making sustainability accessible to anyone with a phone or laptop. -Creating a simple, smooth user experience across different features. -Designing clear, actionable insights that genuinely help reduce environmental impact.

What we learned

-Vision-based AI can extract incredibly useful climate insights from everyday images.

What's next for BioMatrix

-Add AI-powered future simulations (e.g., how your home/neighborhood will look under heat/flood scenarios). -Build a community leaderboard to encourage friendly eco-competition. -Partner with schools & companies to integrate sustainability scoring into real-life programs. -Expand datasets to improve waste classification and carbon estimation accuracy. -Create a mobile app for real-time sustainability nudges and notifications.