ingest and parse PDF files

[RADSr]

I know ( er, I'm pretty sure I know ) that Incorta has the ability to receive a PDF file and work with it - I expect that means to summarize info as in the CFO demo  - which is awesome.   I'd also like to use it to pull data from a table w/in the PDF.

Can I do that w/ Incorta, and if so would it be using data studio, notebooks, something else?

Looking for a pointer to documentation, white papers, videos, whatnot.

[dylanwan]

Yes we have customers who have the same requirement and we use LLM capability to parse the PDF.

We created a custom recipe in the DataStudio for loading the pdf files from cloud storage like S3 or Azure.  LLMs, like Gemini, have a good capability of recognizing text in images.

We can provide this as part of a POC if you are interested.