Recently a colleague of mine came to me with a fun problem to solve. Lawyers get a lot of documents they need to keep organized, could we use AI to help with that? In reality the idea is significantly more niche and focused than the aforementioned one liner but for now let's just focus on that.

I have previously used AWS Textract to pull data from printed forms that were photographed so why not use it here? It is a pretty amazing little tool, depending on your margins it can be a tad pricey though. Currently, depending on how you are using it around $0.25 per page.

Basically the idea would be to parse the documents and to start determining patterns in the documents that we can use to classify them. We might even go as far as to run the extracted text from the documents through an LLM to have it take a stab at classifying the documents.

It is a fun little project. Not sure if it is a green light yet but in the meantime I will have fun dreaming up how all the pieces will work together. Perhaps I will use the new diagramming tool to mock up how it would all look together.

~Cheers

PS: If you are interested in engaging my services as a Web Application Architect check out my Group Coaching Program or 1 on 1 Consulting at schematical.com and check out my FREE eBook 20 Things You Can DoTo Save Money On Your Amazon Web Services Bill Today