There have been a number of folks here who seem to think AI is some kind of antiChrist. I hope to show them that while it may be complex, at the end of the day it relies on plain 9ld plumbing. As AI shifts from experimental projects to core enterprise “connective tissue” (as mentioned in the article’s news sidebar), the underlying infrastructure must become more resilient and less prone to disruptive, “bulldozing” repairs. As I dig deeper into how we can use different models in my projects I find it less and less awe inspiring and more and more spaghetti code like issues arising
It was really none of my business but thank you for explaining this. There are a few here who understand what the article is discussing. The ones who are most qualified probably have not yet seen it. As you are no doubt already aware most of the others here are likely not very familiar with Kubernetes, Proxmox, Docker, or even Virtual Machines.