What it does
The program looks through a client website's catalog and records the products descriptions, feeding it through an AI which scans the description and separates them into quantifiable metrics based on a rubric which an AI can use.
The program then sends these AI-friendly descriptions to a database so that the AI chatbot can look through easier, rather than scanning a potentially AI-unfriendly website. The program also sends the client a score based on the description of how AI-friendly a given product's description is, as well as what can be added to the description so that it can be made so.
How we built it
Using python and an external LLM, we were able to accomplish what we set out to do above.
Challenges we ran into and acomplishments that we're proud of
Our group is not particularly well-versed in coding, so creating a working prototype was quite challenging. However, we managed to push through and create a substantial prototype that we are satisfied with, despite our lack of experience.
Log in or sign up for Devpost to join the conversation.