Automating document processing for a legal services firm

A legal services company employed 50 people in India who spent 60–70% of their time on one manual task: splitting large PDF files into their individual documents and filing each one into the right folder on the company's server.

These were legal documents, so no data could leave the company's server. Any tool that sent the files to an outside service was ruled out.

Using Claude as our development partner, we designed and built a Python application that runs entirely on the company's own server. It splits each PDF into its component documents and files them into the correct folder structure automatically.

AI helped us write the software. The software does the work locally, so the documents never leave the company's environment.