They're a cool team and I like the idea so I hope it works. In our case, for the few million, we could get a hell of a lot more hardware (Epyc 9654 based machines to start with - much better operating cost / compute) so the magic must be in the software.
It is true that because we're a young company, we currently have the initial SKU that we started selling two years ago. Like every hardware company, we'll be releasing new products with new hardware every so often. Totally hear you if the current product doesn't fit your needs though, that's just going to be the case when we're so young, but as we grow and scale, we'll be able to release more variations that could make sense for more people.
There is so much more than just slapping servers in racks when you reach the hundreds of thousands or millions in server hardware. Good control plane software makes a night and day difference.
Admittedly this would have been an exotic for the Bay Area but mundane for elsewhere use-case since we just needed really fast disk and fast compute because we just ran model tuning and backtesting on the machines. 90% utilization, 100G network, 4xNVMe. We only had an 8-member team so much of the management/IT-layer replacement stuff that Oxide enables wouldn't have been as high leverage across the tens of thousands of cores. We almost certainly left a lot of stuff on the table, though.
Also, is your username something interesting? It feels familiar but a quick `echo -n '' | md5sum` didn't yield anything.
