HN user

deflator

120 karma
Posts0
Comments73
View on HN
No posts found.

Very informative article. I was hazy on where NVLink sat in the local inference stack. This makes it very clear: to adapt consumer hardware that lacks multiple full x16 PCIe lanes.

I switched to Codex in November 2025 because I couldn't stand the smarmy, "You're absolutely right" anymore. Codex at the time offered less sycophancy and more pushback when I was actually wrong. I also feel it writes cleaner code since it follows my instructions better to not write unrequested code, not create unnecessary abstractions, and not reinvent any wheels. I use LLMs on a tight-leash in HITL for web development, monitoring traces as it goes. I rarely use it for more than just writing code that I specify and I rarely let it "have it's head" and work unmonitored. I use it on the API plan since I need it to be reliable. No outage issues that I can recall.

First taste is always free! But when the 10 billionth taste continues to be free, you have to start to wonder about your dealer.

More or less our approach here too. They offered to help with our work at a loss for themselves, why not take them up on it? Not necessarily to hurt the AI labs, but just because we'd be foolish not to take something offered at such a discount.

MAI-Thinking-1 2 months ago

Does this mean that work created with it can be copyrighted? Since the courts ruled that the inclusion of pilfered IP was the reason other model's work cannot be copyrighted, I would think so! In that case, this is a completely different beast. It can maybe be used for things that need a durable copyright.

I have always taken an untrusting approach to LLMs for coding, born out of them being absolute trash at first. They're 1000x better now, but I still guide it every step of the way. It limits my speed gains, but I got what I wanted: no more typing out code and looking up patterns to get the exact syntax.