Inspiration
More than twenty thousand pets go missing in Washington each year, leaving heartbroken owners desperately searching for their companions. We connect owners with a broader audience to find their pets. There are many issues with the biggest pet finder websites: 1. Most only contact professional pet finders and don't have listings with all the lost animals that need finding. This limits the amount of people who can search for the pet and leads to lower success rates. 2. The search function is extremely limited and makes it difficult for finders to identify lost pets.
What it does
The user experience is extremely simple: Pet owners report lost pets, and all users can view a listing with all the lost pets. To narrow down their search they can use an AI-powered search engine to identify lost pets.
How we built it
The Lost and Hound webpage is built using HTML and CSS. Three of our members designed and built a front-end UI. We looked at public websites around the web to gain inspiration and chose a warm color scheme and minimalist design. Then, two of our members used HTML and CSS to build individual pages (About, Login, Home, etc.)
The backend was built using Express.js, and the information was stored in MongoDB. The backend also interacted with the OpenAI API and used GPT 3.5 to create a more seamless search experience.
Challenges we ran into
Our biggest challenge was brainstorming plausible and achievable yet unique ideas. This would set us back a lot as the time that could have been spent coding was spent brainstorming. Another major challenge that lowered our productivity was the wide variety of skill sets in our team. Only three of us had experience with web development, and only one of them knew how to create a backend.
Accomplishments that we're proud of
We’re proud of being able to use our different skill sets to our advantage. Everybody had a unique role, like graphic design, creating different pages on the front end (login page, footer, about page), and creating a backend. Also, we’re proud that we developed a fully functional yet visually pleasing site in an extremely limited amount of time.
Finally, we were able to able to incorporate AI to make-sure our site is future-proof and extremely easy to use, raising the success rate of finding a furry family member.
What we learned
- How to use the OpenAI API. Prior to this none of us had used AI before; this hackathon was a learning curve that helped us improve our understanding of AI.
- Improved our presentation and teamwork skills.
- Gained experience building websites. Only two of us have made a full website before, so it was a great learning for all of us.
What's next for Lost and Hound
Unfortunately, there are limitations to how much we can get done in 1.5 days. Features that we would love to add to Lost and Hound include:
- Training the AI for better search algorithms
- Multiple pages to organize listings once listings increase
- Better location features with a real-time map. It will have pins to mark every city where a pet is lost
- Dashboard for users that contains all their postings and allows them to edit or remove their lost pet reports
Log in or sign up for Devpost to join the conversation.