HackerLangs
TopNewTrendsCommentsPastAskShowJobs

nullbio

249 karmajoined vor 10 Jahren

Submissions

Ubuntu releases Workshops: Sandboxed dev environments in a single command

discourse.ubuntu.com
6 points·by nullbio·letzten Monat·0 comments

Supply Chain Attack Targets Laravel-Lang Packages with Credential Stealer

aikido.dev
2 points·by nullbio·vor 2 Monaten·1 comments

Hardware LLM Taalas Reaches >14,000 TPS on Llama 3.1 8B

taalas.com
3 points·by nullbio·vor 2 Monaten·1 comments

Ask HN: Supplements for Memory That Work?

3 points·by nullbio·vor 9 Monaten·8 comments

comments

nullbio
·vor 19 Stunden·discuss
[dead]
nullbio
·vor 19 Stunden·discuss
I want my model to help me build up its own infrastructure that instills it with the sort of constraints I want for my project, rather than have it behave generically and automatically for everything.

It should follow instructions incredibly well while inferring contradictions or gaps in logic and surfacing those to the user as suggestions for improvements and persistence.

I really hate how Claude just assumes you want to do X/Y/Z and goes off and breaks everything and you're constantly screaming at it STOP DOING THAT. Instead, it should just do the minimal things while building its own guidance along the way in a persisted memory, like, 'would you like me to do X, now, and in the future?' etc.
nullbio
·vor 19 Stunden·discuss
Agreed that they shouldn't be writing vulnerable code to begin with. You'd think the models would be trained to know when they are working on something that has security implications, and to validate the security of what they're building, as they're building it.
nullbio
·vor 19 Stunden·discuss
This is already happening at scale by the social media feed algorithms. We don't need generated content to accomplish this. In a sea of user created content, plenty of it is already at peak activation.
nullbio
·vorgestern·discuss
Someone has to know.

Would be nice if an insider would drop some hints so that the open-source space could make some good progress.
nullbio
·vorgestern·discuss
I disagree. Fluid natural conversational AI is far more productive than any other interface for working with LLMs. Although I suppose you could make the argument that it should be more... Robotic like. Like in StarTrek. Which, is honestly probably better for work, too. A "get shit done" mode, of pure, cold, efficiency.
nullbio
·vorgestern·discuss
It took him a year to build. So yeah, obviously if someone spends a year working on something with an LLM they can produce a good product.

The slop we're seeing from people using AI is because they pump it out in a month or two and then call it a day.
nullbio
·vor 3 Tagen·discuss
https://clawd.rip here's a start.
nullbio
·vor 3 Tagen·discuss
I've used it extensively.
nullbio
·vor 3 Tagen·discuss
Yes, but better.
nullbio
·vor 3 Tagen·discuss
Everything is vibecoded now, better get used to it.

Aside from that, the quality of the project is great. Vibecoded doesn't automatically mean bad.
nullbio
·vor 3 Tagen·discuss
It's better than tmux. I've used both extensively, there's no reason to use tmux again.
nullbio
·vor 3 Tagen·discuss
It's just empty hype. It's mildly better than Opus 4.8 at best.
nullbio
·vor 3 Tagen·discuss
Cancel your subs. Anthropic does not value or respect its customers at all.
nullbio
·vor 3 Tagen·discuss
Who cares, that doesn't have anything to do with the way Anthropic behaves. Not everything needs to be turned into something about Trump.
nullbio
·vor 3 Tagen·discuss
I really hope people cancel their subs once GPT 5.6 drops. This company deserves to sink.
nullbio
·vor 3 Tagen·discuss
Agreed. They've been doing this sort of thing since the beginning and it's infuriating that people eat it up. I really despise this company.
nullbio
·vor 3 Tagen·discuss
Next up: Thoughts from an OnlyFans model.
nullbio
·vor 4 Tagen·discuss
Can promise you a re-install does nothing for your privacy. Plenty of IDs are embedded in the hardware.
nullbio
·vor 4 Tagen·discuss
You're correct. It's just the latent space of the transformation. Nothing magical here, they're effectively breakpointing the model at the layer level and switching the activations in real time. It's pseudo-scientific bullshit designed to push a narrative.