Inspiration
The three of us shared the frustration of watching our families struggle to navigate the complex healthcare claims process. These experiences inspired us to create a tool that streamlines the process, helping people focus on getting the care they need instead of deciphering confusing insurance policies.
What it does
"Ask a Duck!" simplifies healthcare navigation. Users select their healthcare plan and describe the coverage or treatment they’re seeking. The "Duck" then reads their specific plan's coverage and benefits declaration, providing a clear answer on whether the care is covered and outlining any steps needed to access that coverage.
How we built it
We chose Flask for our backend because of our team's familiarity with Python. The website is hosted locally on a Windows server, with Packetriot enabling internet access via secure tunneling. For domain registration, we used GoDaddy. Our frontend is built with HTML, CSS, and JavaScript, leveraging tools like ChatGPT for design inspiration and development efficiency.
Challenges we ran into
This was our first time self-hosting a website, and we encountered significant hurdles, especially with domain registration, DNS configuration, and managing the hosting environment.
Accomplishments that we're proud of
We successfully integrated OpenAI's ChatGPT using developer tokens and Google Translate through API calls to enhance the app's functionality. These integrations help deliver seamless, user-friendly interactions and multilingual support.
What we learned
This project was a deep dive into web development. We gained hands-on experience with self-hosting, including domain registration and DNS settings. We honed our frontend skills in HTML, CSS, and JavaScript design and explored Flask’s capabilities for building robust web apps and making API calls.
What's next for Ask a Duck!
Our next steps include adding support for more healthcare plans to broaden the app's accessibility. We also aim to implement a locally hosted language model (LLM) trained on real-world case data. This enhancement will improve accuracy and provide realistic predictions beyond what is apparent in standard insurance documents.
Log in or sign up for Devpost to join the conversation.