AFAIK this is mostly incorrect. First the crawler archive URLs to files in WARC format, which are uploaded to archive.org, which then imports the files into the Wayback Machine. For ArchiveBot jobs, you can find the WARC files for the completed jobs for any domain from the ArchiveBot completed jobs website. For other types of jobs, including custom ones like DPReview.com the process is similar but involves more machines due to their more distributed crawling process. It all ends up in the Internet Archive Wayback Machine though.
https://archive.fart.website/archivebot/viewer/ http://archiveteam.org/index.php?title=ArchiveBot