Large language models (LLM) have revolutionalized the way we access information. A quick trip to ChatGPT which is a Generative Artificial Intelligence (GenAI) app built on LLM produces results based on the vast data it was trained on. The question that arises is that of proprietary data which it was not trained on. In other words, how do you gain insights from a private PDF file, videos or articles that are not publicly available on the internet?
My project which is an app I call "MyAI" is a custom solution to one of the challenges mentioned
above. It allow users upload a pdf file and chat about this file to gain insights on it. In the
screenshot below, I uploaded a pdf file that talks on strategies for reading research papers,
and went on to ask questions about this file in an interactive Q&A pattern.
-
Add link to deployed project website. At the moment here is my landing page which contains link to a youtube video where I demonstrated a live usage of the app. Project Demo
-
Improve functionality to include processing of video files
I worked on this project alone. My aim was to test my skills in building a full stack app. Lets connect on linkedin