Hearth

"Building Connections, Creating Homes: Empowering communities by pairing those in need of affordable housing with homeowners seeking support."

Comment

Inspiration

The inspiration for Hearth came from two big issues: skyrocketing housing costs and the need for accessible services like nannies, tutors, or companions for the elderly. We thought, why not connect the two? Hearth is our modern, community-driven solution that matches homeowners with people seeking affordable housing in exchange for valuable services—making homes and hearts a little fuller, one match at a time!

What it does

Homeowners can create profiles detailing their available living spaces and the assistance they need, such as nannying, tutoring, elderly companionship, or housekeeping. They also specify the home’s location, and other important information, including amenities and the duration of stay. On the other side, service providers—such as students, caregivers, or workers—can sign up by creating profiles that highlight their skills, services offered, preferred locations, and availability. Hearth solves two challenges at once - providing affordable housing and fulfilling essential service needs.

How we built it

We used React to create a dynamic and interactive user interface. We also utilized TailwindCSS for our styling needs, making it easier to design a clean and responsive layout. We also utilized Next.js for server-side rendering and routing. However, we also used HTML and CSS formatting. In our attempts to integrate AI, we mainly used python and looked into possible models/APIs like those from Hugging Face and OpenAI. Despite our efforts, these AI systems didn’t function as expected within our project. As a result, we were unable to achieve the AI functionality we had aimed for.

Challenges we ran into

We ran into a lot of time constraints while creating our project, especially due to the fact that many of us had to learn all of the technologies we were working with for the first time! One of our biggest challenges was diving into front-end development. None of us had much experience with React, so merging our basic HTML and CSS experience with this new framework was a bit of a learning curve. We spent a lot of time figuring out how to adapt our previous web design skills to React’s component-based system. Another one of the biggest challenges we faced was integrating AI into our project, and, to be honest, we didn’t quite get it working the way we wanted. We wanted to build an AI-driven matching system to pair service providers with homeowners based on preferences, but it turned out to be a lot harder than expected. We spent a lot longer than planned fine-tuning the algorithms, trying to make sure they were accurate and useful, but it didn’t come together in time.

Accomplishments that we're proud of

We all learned React, TailwindCSS, and Next.js, which we all had little to no experience with. Several of us used GitHub to collaborate, which most of us have not done before. In addition, we were able to create some beautiful mockups in Figma! We’re proud of the final, polished look of the website. Finally, we gained knowledge about the OpenAI API and utilizing and finetuning GPT models in our code. Although our AI matching system didn’t ultimately work, we made a lot of progress in understanding and implementing AI systems and API integration.

What we learned

We learned that it’s important to plan out small, manageable steps, and figure out what everyone’s best at to have them work on it. In addition, we found that it’s important to get the functionality working before we code the front end of a website.

What's next for Hearth

In the future, we aim to fully implement Hearth's smart AI matching system. While we made significant progress during this hackathon, we weren't able to get the system fully working. However, our goal was to have the AI pair homeowners with service providers by analyzing factors like location, skills, availability, and preferences. Once the system is working, it will improve the user experience by making seamless and accurate matches based on personalized data, ensuring both homeowners and service providers are satisfied with their pairing. We’re also excited to flesh out our database system by furthering the Explore Page’s ability to pull from the housing database that the List A Hearth page writes to! In addition, we’d love to implement a background check program with a third party such as Checkr to ensure that this program was safe and secure.

Built With

Share this project:

Updates