Inspiration
Toiling through long pdf files and looking for one piece of information can be frustrating, and since gpt is quick and relatively string at data extraction, I felt it would be an excellent tool to provide to consumers.
What it does
Asks for keywords and a pdf file and then returns the results of gpt-3.5 looking for the keywords
How we built it
I use Llama Index to process given pdfs as well as request the same query from the OpenAI API
Challenges we ran into
I am relatively new to programming in Python, so some Python-related bugs were time-consuming due to my lack of experience. Also, some of the Llama index example code would not work correctly when I tried to implement it so I had to work with mentors to learn the actual methods.
Accomplishments that we're proud of
Learning a new framework and utilizing it.
What we learned
I learned much more about utilizing AI as well as its limitations.
What's next for PDF Interpreter
Implementing more defensive programming practices and multiple file comparisons.
Log in or sign up for Devpost to join the conversation.