I think part of the issue is that Azure has been struggling to reliably provision Linux VMs. Whether that's due to increased load, poor operational execution, or a combination of them, it's hard for anyone on the outside to know.
> It's because your brain is predicting what they will say while they speak, and processing an answer at the same time. It's also why when they say what you didn't expect, you say, "what?" and then answer half a second later, when your brain corrects.
that's super interesting. do you know of any resources to learn more about this phenomenon?
prompt: what's the term for financial moves where you keep finding new investors to pay off the previous ones
GPT-4o mini: The term for financial moves where new investors are continually recruited to pay off previous ones is often referred to as a "Ponzi scheme." Another similar term is "pyramid scheme," where returns are paid to earlier investors from the contributions of newer investors, but with a structure that typically requires participants to recruit others to earn returns. Both schemes are unsustainable and illegal.
I always find these discussions about AWS NAT gateways interesting because I recall way back in the day, before AWS had a manages NAT gateway, the recommendation was to roll your own anyway. Or at least that's what I heard. I took an ACloud Guru course and one of the first ec2 lessons was to create a simple NAT gateway in your VPC so that your other instances could reach the Internet.
It's more than that. They have both their own completion model and now agentic one. It's not a basic fine-tune, because it's faster than anything else available out there, so there's something interesting in the architecture itself.
yeah its not fair to call it a finetune because finetune carries connotation of "there wasnt that much extra compute and data added". RLFT has a lot more added to it as Sasha alluded in his talk https://www.youtube.com/watch?v=md8D8eNj5JM - the x axis is log scale, think about that
the framing here is more about "why would you start from random weights when perfectly good starting weights exist" https://www.latent.space/p/fastai
I'd argue this is an issue not just with "non-tech" folks, but even engineers who don't have experience with Prometheus and other time-series databases. Learning promql always seemed like a hard thing to ask of other engineers. Grafana has made it easier to explore and build queries over time, but there are still quirks and nuances that can be difficult to explain to people whose role doesn't typically involve scouring through metrics.
reply