I ran out of time before I could properly finish the Project Story for the submission, so I'm adding it here:
Inspiration
I worked as a translator during my senior year at Arizona State University, and I think I've always wanted to go back to that area and build something to help people who are not native English speakers.
What it does
It takes a document, like appointment reminders or Medicaid information, translates it into Spanish, then highlights key relevant information, such as dates, contact information, and purpose of the communications. It also flags messages that could potentially be malicious, showing the user what the red flags are.
How we built it
I used the skills given to me for this hackathon and built the whole thing with Claude, with the LLM taking care of the coding elements and me the ideas and decision making.
Challenges we ran into
The biggest challenge was coming up with the initial idea for what to build. How do you build something that uses AI, without it being a worse version of what ChatGPT or Claude can already do? I think the approach that I took, to focus less on the model capabilities and more on ensuring consistent outputs and taking into account an audience that's not comfortable with prompting, addresses that challenge.
Accomplishments that we're proud of
Being able to complete an app like this in such a short amount of time. It's a first for me, and I'm actually pretty proud of how it turned out.
What we learned
I learned about skills, which I never used before, and how they can streamline the development process.
What's next for Communications Interpreter
I'll generate more test cases and look into expanding the scope to incorporate other languages.
Log in or sign up for Devpost to join the conversation.