Solution 1:
// loading bytes from a file is very easy in C#. The built in System.IO.File.ReadAll* methods take care of making sure every byte is read properly.
// note that for Linux, you will not need the c: part
// just swap out the example folder here with your actual full file path
string pdfFilePath = "c:/pdfdocuments/myfile.pdf";
byte[] bytes = System.IO.File.ReadAllBytes(pdfFilePath);
// munge bytes with whatever pdf software you want, i.e. http://sourceforge.net/projects/itextsharp/
// bytes = MungePdfBytes(bytes); // MungePdfBytes is your custom method to change the PDF data
// ...
// make sure to cleanup after yourself
// and save back - System.IO.File.WriteAll* makes sure all bytes are written properly - this will overwrite the file, if you don't want that, change the path here to something else
System.IO.File.WriteAllBytes(pdfFilePath, bytes);