Sendgrid seems to have a bug in the Parse api. The fields are inconsistently encoded. According to the docs, they all should be UTF-8, but they are not.
I sent an email to the parse endpoint with the subject and the body containing: 日本国.
The text one below is incorrect (not utf-8)
Here is a capture of the data on the wire:
5a 59 0d 0a 43 6f 6e 74 65 6e 74 2d 44 69 73 70 ZY..Content-Disp
6f 73 69 74 69 6f 6e 3a 20 66 6f 72 6d 2d 64 61 osition: form-da