Skip to content

jcbendernh/DatabricksPDF

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 

Repository files navigation

DatabricksPDF

The purpose of this repo is to showcase how you can take a PDF document such as a resume and using the ai_parse_document function, you can extract structured content from unstructured documents into a Delta table. This is showcased in the Parse Documents to Silver notebook.

Using the ai_query function, you can further curate the data into separate fields via LLMs. This is showcased in the PDF Gold notebook.

Below is a reference architecture.

Architecture

For more on this overall topic, check out PDFs to Production: Announcing state-of-the-art document intelligence on Databricks

This architecture was for a Proof of Concept and is not considered production level code. For something more production ready, check out the link at the bottom of the PDFs to Production article referenced above.

About

DatabricksPDF is a proof-of-concept repo showing how to extract structured data from unstructured PDF documents (e.g., a resume) into a Delta table using Databricks SQL’s ai_parse_document. It then demonstrates curating that extracted content into “gold” fields using ai_query/LLMs, with example notebooks and a reference architecture diagram

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages