Build a tool to extract data from PDF engineering drawings with set title blocks

I have about a thousand files with data I was hoping to extract. At a minimum, I would like to see if someone can help extract the data from these files for the title block information. The title block is generally consistent in format within a given set of files with a small number of exceptions only, so it should be fairly managable with some tweaking. Secondly, there are some additional tables with item numbers on them and relevant attributes, and ideally, if that can be extracted, that will be awesome as well. But these data are highly variable, so it might be difficult for one to extract these data, and I am not sure how one may be able to achieve such without in depth knowledge about the drawings. Therefore, this RFQ focuses on the first part, and the second part should be a separate offering if you think you can do it - which is to extract item numbers and their relevant attributes that were organized in an unstructured format. Examples of title blocks can be found on google by searching engineering title blocks. Detailed formats will be provided when the project gets awarded. There are approximately 500-1000 drawings that will need indexing for this project. Please bid accordingly based on that figure.