Upon investigation, it appears the cause of this issue was an unexpected update made by our hosting provider. Errors have since subsided to normal levels as of 5:15pm Central time.
Engineers are currently investigating an increased error rate on our backends.