Inspiration
Our inspiration for creating this platform stems from a deep-seated desire to address the pressing environmental challenges posed by inefficient waste management practices. Witnessing the staggering volumes of waste accumulating in landfills worldwide and the detrimental effects on ecosystems and communities fueled our passion for change. We are driven by a vision of a world where waste is viewed not as a problem, but as a valuable resource to be managed sustainably. By harnessing the power of technology and innovation, we aim to empower individuals and organizations to play an active role in shaping a more circular and environmentally conscious future.
What it does
WasteWise is an AI powered waste management platform that is dedicated to manage, recycle and minimize waste. It utilized a vision-based LLMs to accurately identify various types of waste and provide users with tailored recommendations on how to properly recycle them. The features of this platform are:
- AI Image Detector: WasteWise's AI image detector employs machine learning algorithms to instantly recognize and categorize various types of waste from user-submitted images.
- Generative Search Functionality: WasteWise's intuitive search functionality allows users to easily identify and learn about specific types of waste. Users can input keywords or descriptions of the waste they have, and WasteWise provides step-by-step instructions on the most environmentally friendly disposal methods, empowering users to make informed decisions about their waste.
- Service Providers Recommendation: WasteWise features a comprehensive directory of organizations offering waste management services, including waste collection, recycling, composting, and more. Users can explore and connect with service providers in their area, facilitating access to sustainable waste management solutions and promoting collaboration between users and service providers.
How we built it
WasteWise is architected with React.js and Tailwind CSS for the frontend, providing a responsive and scalable user interface design with maintainable code structures. The backend infrastructure is constructed using Node.js, offering asynchronous event-driven architecture for high performance and scalability, complemented by MongoDB as the NoSQL database for efficient data storage and retrieval.
Our machine learning models are meticulously engineered in Python, harnessing the capabilities of TensorFlow, NLTK, NumPy, Pandas, and other essential libraries for data manipulation, natural language processing, and machine learning tasks. Additionally, we have integrated LLaMA as the language model, facilitating advanced text generation and comprehension capabilities, alongside incorporating OpenAI's GPT-4-Preview APIs to access cutting-edge natural language processing functionalities.
Challenges we ran into
- Coming up with an innovative practical idea that is built on existing tools users are familiar with to help with adoption.
- Originally, we planned to make a mobile app but since none of us had experience with building a mobile app, we had to switch to making a web platform
- We ran into some issues using React.js, but we were able to solve it together
Accomplishments that we're proud of
- Collaborating together as a team (even though we were in 3 different continents)
- Completing the project within the given time frame
- Being able to implement most of the technical features
What we learned
- Efficient time management and collaboration
- Prioritization of important tasks
- Problem-solving abilities
- Prototyping and creating an MVP
What's next for WasteWise
While we have a solid foundation on our platform, we are committed to continuous improvement and innovation. Here are some of the features we plan to add:
- Expanded service provider directory
- International coverage
- Multilingual support
- Advanced search filters
- Integration with sustainability metrics
Log in or sign up for Devpost to join the conversation.