Hacker Newsnew | past | comments | ask | show | jobs | submit | tmikaeld's commentslogin

Why where they using kubernetes for this at all, it’s attack surface and config complexity is too large for anything that needs airgapped security. Yes, I know kubernetes used for large production deployments for web apps and that is my point, it’s not made for securing AI agents with insider access, it’s meant for securing outside access.

It's just cached by CF Tunnel

How much does it cost per month, which provider and what do you get out of it?

Hmm? I'm running the models locally... 2x Sparks consume ~150W at peak, and they usually spend more time waiting on results from whatever task they're working on, so I imagine that the contribution to my electricity bill is maybe a dollar/mo or less. Though, of course, each Spark was $4000, so the total I've spent is equivalent to several years of the maximum tier for most cloud model susbcriptions.

What I get out of it is the ability to hand login credentials to my other computers to manage their updates, bug fixes etc. Eg. After updating my proxmox server, the nvme drive kept dying. Was able to let my local AI in to figure out and fix what was wrong (known issue). A cloud-based AI could've done it too, but I don't want to be sending internal passwords out of my network like that.

Plus, the ability to freely delegate tasks or exploration of things cloud models generally avoid. For example, I draw as a hobby, and when I'm struggling with a pose but can't quite figure out what I'm missing, I pass it into a VLM for advice, but Claude etc get unnecessarily cautious because they interpret an anatomical sketch as a naked person.


Those posts are so many now that I’m inclined to believe they’re all written by AI bots..


I’m on X for tech interest, new hardware, robotics, coding. If X didn’t allow blocking words/phrases, I could not use it! And yes, my blocklist is huge..


That’s really impressive! I’d say even better than fable managed


To be fair, I misspoke, the NVFP4 quant I used for that test runs on 8 cards. Still...!


TOPS seems to be low though, so fill rate is probably ~10x slower than a 4090.


Not only is it good - but the maintainer is active and really nice, helped me fix two bugs straight off the bat!


The problem with lovable, from someone with insider knowledge, is that many of the apps existed even before appearing there and where ported to the platform to ride the hype wave.


My problem with DS flash/pro is that they don’t push back on obvious bullshit, both irl and code [0] but it’s a great implementer workhorse if you give it _very_ detailed specs.

[0] https://petergpt.github.io/bullshit-benchmark/viewer/index.v...


omp.sh lets you enable "advisor" model that watches the session. Works very well for keeping DS on track.


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

Search: