While this isn't a "doctor" implemented with GPT, it certainly could act as one if someone indexed enough medical related documents with it, or physics documents, or philosophy documents. We certainly don't recommend anyone implement a medical doctor with it, however.
The intent here was to build something that was simple to understand, didn't use an opinionated framework, and leveraged the idea of semantic knowledge graphs to manipulate the vector space, with the intent of increasing prompt relevancy.
If you find this valuable, I'd love some feedback on it. There are plenty of PDF indexing solutions out there for GPT, but many of them have limitations with document size and relevancy. It is my hope to continue working on this concept to make all PDFs accessible to LLMs, hosted or otherwise.
The intent here was to build something that was simple to understand, didn't use an opinionated framework, and leveraged the idea of semantic knowledge graphs to manipulate the vector space, with the intent of increasing prompt relevancy.
If you find this valuable, I'd love some feedback on it. There are plenty of PDF indexing solutions out there for GPT, but many of them have limitations with document size and relevancy. It is my hope to continue working on this concept to make all PDFs accessible to LLMs, hosted or otherwise.
Thank you!
Kord Campbell