Inspiration
While online document services can store text, no other tool can digest information and organize a story's structure the way GPTree does.
What it does
GPTree takes text input which is then analyzed by OpenAI's Chat GPT. The initial idea was to have this input specifically analyzed according to certain lenses that speak closely to readers. GPT would then return ideas to expand the storyline and the user could accept or decline these additions.
How we built it
The web app is built using React, and the backend is powered on a local server. The API calls are directly to OpenAI
Challenges we ran into
It took some time to get calls to properly hit OpenAI's servers, but once that was figured out the project could be expanded.
Accomplishments that we're proud of
This was one of my first times using an API and developing a full-stack application. I believe I did a good job.
What we learned
I learned that I can do more than I think. I had little hope for myself coming into this hackathon, and although there were some issues with my team, I was able to still pull through and make an effort.
What's next for GPTree
The next steps for GPTree would be editing the response that GPT gives, and asking the AI model to look for specific plot holes, storyline developments, etc. From there I would implement a tree visual to track the progress of the story as it goes.
Log in or sign up for Devpost to join the conversation.