PDF Extract callback notifier isn't being called

I've been testing the pdf extract api but I can't get the callback to work, I've tried numerous different things but it never seems to be called.

The callback url I have is hosted on aws api gateway and publicly available without any auth, I've been able to call it from other servers on the internet so there's no way that Adobe shouldn't be able to access it.

Here's a minimal command I'm using to request extraction,

curl https://pdf-services-ew1.adobe.io/operation/extractpdf --header "Authorization: $ADOBE_TOKEN" --header "x-api-key: $ADOBE_ID" --header "Content-Type: application/json" -X POST -d "{
      \"input\": {
        \"uri\": \"$PRESIGNED_PDF_URL\"
      }, \"params\": {},
      \"notifiers\": [
        {
          \"type\": \"CALLBACK\",
          \"data\": {
            \"url\": \"$CALLBACK_URL\"
          }
        }
      ]
    }"

The requests are accepted successfully and do return a Location header that I can poll to see that it's in progress and then eventual completion with the download link, but I just never receive a callback at any stage. What am I doing wrong?