Sounds like a you problem. I’m all for people investigating the boundaries of model capability - if you take that as a personal attack, you’re going to have a bad time over the next few years.

While I suspect "hot take" is apt, this wasn't exactly anti-AI. Rather, the author is advocating for their particular way of doing genAI output parsing (as opposed to constrained decoding structured output, they advocate unconstrained decoding with a permissive parsing framework)

How is this an "anti AI hot take"? It's discussing using one type of LLM output versus another...