Oy. Sadly, I suspect anybody who has worked in dev knows the outlines of this story: Management promising deliverables with unrealistic timelines and functionality; team leads with such poor internal product knowledge they don't recognize catastrophic issues until it's too late; poorly-trained devs squeezed to push out unscalable kludgy insecure nightmares "for now" that just never get addressed; systems gradually becoming a Jenga tower of manual workarounds and undocumented slapdash quick fixes... Ugh. Nightmare fuel.

A very entertaining read. I'm completely unsurprised, based on my experience trying to use Azure. Garbage platform.

And now they're hacking on that shitshow with CoPilot. Jesus wept.

TLDR: A former Azure Core engineer details how Microsoft risked losing OpenAI and government trust due to complacent decisions, including a plan to port half of Windows to a tiny chip, which the engineer deemed impossible. This mismanagement, among other issues, potentially cost Microsoft a trillion dollars in market capitalization and led to wasted engineering efforts.

All articles in one page

lmfao

Article proposes that losing trust from OpenAI and the US Government are the causes of their woes, when truly the opposite is the case. OpenAI and the US Government are the reason why everyone hates MacroSlop and all of their products are rapidly degrading.1

having been part of the Windows team since 1/1/2013

If he worked on Windows 11 he's the part of the problem.

As a former Microsoft employee the eroded trust for Azure came from it's use for a genocide. Not really much more you need to know. If a company is willing to provide cloud and AI infrastructure for the slaughter of civilians and the specific targeting and bombing of journalist, well, it might not be the most trustworthy service for your baked goods website.