@slew no way! Well I actually have questions for you 😅

How hard was it to adapt your work to run on it? I haven't done work on a large AMD GPU cluster before. I assume mainstream ML stuff would be a pain?