An easier approach IMO would be to create an amazon lambda script to process the pdf and make it run when you upload a file to a bucket in amazon s3.
It's easy to do and you don't need to mess with anything more than two simple API.
It's easy to do and you don't need to mess with anything more than two simple API.