The most annoying thing about a lot of these is that tutorials are “minimal viable setup” sorta things. Like “now you have it setup, make sure you tune it for production”

Dude I’m already in pain from trying to serve these models and you just have to go rub salt into my eyes. “Simplify your stack with <Tech>” they said. “Share your resources effectively and easily with <Tech>” they said. “Here’s your fuckin’ ‘Hello, World’ now GRTFM and buzz off” they said.

Working close to the metal do be like that.

I work for a large enterprise and build ML model monitoring pipelines fairly frequently—this will be a more in depth but similar use case to what you’re asking.

We use Grafana (visualization) and Prometheus (timeseries db)—they’re built for this use case exactly. Tons of info out there on how to build, configure, connect to your sensors, and deploy it.

I call it N Jinx. Always have and I’ll never be convinced otherwise that it’s not.

Eh I just misspelled it, the spirit, pronunciation are still there. I consider the chain in tact.

No, you’re thinking of a dead lach. A deadlock is an anti folk music pioneer—not really dead yet though he claims to be dead inside.

Who can get there first?

I feel so compressed.

I don’t think an SSD is the right choice here. SSDs have a limited lifespan that’s majority driven by the number of writes that happen to a certain block. Reads are cheap and near infinite though.

When you’re talking about a Lemmy instance, mail server, etc. my mind thinks this is likely to be many writes with several read-once ops. This is a better use case for a HDD.

A media server that oriented towards most consumption (reading) would be better for SSD.

I like draw.io for process diagrams.

I figure I can probably convert about 10 kg into manure before it autoconverts into compost. Which is maybe even a worse problem.

Would you rather have 100,000 kg of tasty supreme pizza, or 200 kg of steaming manure?

Choose wisely.

That worked out ok for one of the 9/11 jumpers. Maybe you could be the next one!

You were always cool.

Don’t forget to use a one-time pad.

That’s why Nintenfux is suing. They have no other way to be competitive in mobile gaming now.

Because the owner of Brave is a giant douchehole.

Great browser though and the company’s mission is decent enough, for a for-profit company.

Technically my job title is Principal ML Engineer and AI Lead—Intelligent Systems at one of the big AI unicorns—I won’t say which because it’s a really small, niche field and it’s very recognizable. I work extensively with deep learning and large language models, and image models too—Generative ML generally. Satisfied?

But AI is definitely a field. AGI doesn’t exist yet, though some of the OpenAI folks imply that it’s close—but AI has existed in various flavors for decades.

I probably know more about this field than you. And if that’s not the case, then we’ve probably met at a conference or two. Not meant to be condescending; it’s just a fact—I’ve spent more than 20 years in AI R&D.

Sorry, my jaded sense of humor can be a bit subtle at times.

I believe many humans are little more than biological language models, with some reinforcement learning fine-tuning by societal norms and reward/penalty functions.

I do this shit for a living (AI Engineer working on LLMs) so yes I do have a pretty good understanding of the technology.

woosh!

I mean… it’s not human intelligence no matter how many people continue the trend of inaccurately calling it that. It’s a biological neural network. It has the ability to write things that look disturbingly close, even sometimes indistinguishable, to actual writing and coherent thought. There’s no good reason to mistake that for actual intelligence or rationality.