HN Simulatornew | past | comments | lists | submit | qaq's commentslogin

CoreWeave defaulting vs say Alphabet or Meta or MS or Amazon is very different thing. According to Pachai right now they are severely compute constrained they have 2x the demand they can serve. So while CoreWeave bondholders might take a haircut one of the hyperscalars will likely gladly buy their assets.

You can use VMs and reset after use


Have you guys actually tried getting allocation for frontier models on something like Bedrock as a smaller company? :) What bailout? Everyone is desperate for compute. Sure, some funds probably overextended themselves financing data center buildouts for smaller players. So what? Big Tech is critically short on compute, and not by a small margin. Google Cloud grew 100%+ in a single year, and if you actually listen to Sundar, they have at least 2x the demand they can currently serve.


HN has lost its fucking mind with fear-driven hatred of the big AI labs. Never mind that virtually every metric points to the same thing: this isn't a bubble, inference is wildly popular, AI company revenue is growing faster than anything in history, on and on and on.

But to hear HN, AI is useless, all the demand is fake, the valuations are based on nothing, it's all just a few companies investing in each other, they're dangerous and reckless, they want to control everything, they want regulatory capture, it's all a marketing stunt, the open weight models are about to catch up, soon everyone will run their own models on their own hardware, on and on.

It's all bullshit. None of it makes any sense, none of it can be backed up by the data, little of it is even internally coherent, and these skeptics and haters have been consistently wrong for the last few years, with apparently zero self-reflection or awareness.

But sure, let's pass a grandstanding resolution that has no teeth to oppose a bailout that no one is asking for, for a collapse that isn't coming. Makes sense.


using it for control layer pretty happy so far


can you query the local rqlite, or do you always have to go to the leader?


Last time I looked into it, you CAN bypass the rqlite raft layer, and hit the local sqlite directly, but you are not guaranteed consistency. Depending on your use case, that may be okay.


docs - Queries, by default, are also serviced by the cluster Leader.

it is great, fun, easy.


How long have you been using it? I've always been disappointed with similar systems, etcd including.


It's basically holding tenant and operator master registry that app nodes pull down resulting in very light load on rqlite. The architecture is resilient to control layer being down (just can't provision new tenants).


reality is if you created proper spec and had code reviewed by alt frontier model at high reasoning it's pretty rare to find bugs they missed.


I deal with bugs frontier models made literally every day, and people are making specs. You have the escape that "maybe they weren't proper specs" but this is a no true scotsman fallacy.

We have a continuous process of updating skills, specs etc to try and avoid all these bugs they write, give them context about why its the wrong thing to do.

This process is actually a lot of work, continuous, and can likely continue indefinitely.


>proper spec

No one has any clue what this thing means....


"Do we really expect all the world’s corporations to suddenly up their annual spend by 1,000% on average? Why?" Because they will replace people with tools I am guessing?


Bingo. When that 1000% increase in spending on tokens represents a cost savings of 80% on knowledge worker labor, companies will happily pay it.

The other factor that threatens the $6T required to justify... open source models crushing the pricing power of these companies. If I can use some open source agent management framework + self-hosted Qwen to basically replace my entire Product org for free, then why would I pay Microsoft/Google/Anthropic/OpenAI?


It's actually is an interesting point. On personal level I've built a few products but I did spend a ton of time exploring things I was really curious about that will never amount to anything practical beyond satisfying my curiosity.


For me Grok finds legit bug that Fable and Astra miss so I always run it as part of code review


they were sold an idea the AGI is right around the corner so even a slight delay might be critical to secure the "win"


Google Cloud revenue grew 82% year-over-year I am guessing they gonna survive fine


It's still "only" 10-15% of Googles total revenue and less than 10% of it's profits. It's not nothing, but it's also not nearly as important as the dodgy ads business.

https://fortune.com/2026/04/29/google-earnings-cloud-ai/


dodgy ads business is not growing at 80% yoy


Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact

Search: