News for hackers. Stuff that matters.

blinded

Re:Anthropic orders AI tool Claude to stop remaking video games on Saturday October 31, 2026 @05:28PM (#50320419)
right into the arms of open models and pi.dev
Re:Official Claude Code extension compromise on Friday October 30, 2026 @08:25PM (#50311959)
You guys open your editors?

This is a joke (neovim user here).

Dude are they trying to get dupped like ArtCraft with Photoshop?
Re:Discord Just Couldn't Resist... [video] on Saturday October 10, 2026 @10:41PM (#50039320)
Its a good product for voice and video chat with friends when gaming. But holy is it bad at chat, the threads ux is weak as shit and the age verification stuff is just odd.

Not a fan of the gambling stuff either.

Re:Pi 1.0 on Thursday October 01, 2026 @09:44PM (#49929049)
use it daily, its great.
Good luck getting into it
Re:Spec Driven Development in 14mins on Monday September 28, 2026 @05:58PM (#49885023)
spec driven dev with good search is the only way I have been able to get consistently good output from a cheaper model
hard agree.
Re:Nice Play Anthropic on Sunday September 27, 2026 @12:46AM (#49863434)
Its great for sure, if you have the hw to run it! If not you're left spinning up GPU instance (which I do!) but for the average bear they just something that works!
Can't image working there, between this drama and the wordpress tiff last year.
Re:Ask HN: Which model do you use for work? on Friday September 25, 2026 @03:26PM (#49848820)
luna for day to day tasks. opus for design / specs.
Re:AI Infra Is Nothing Like the Classic Cloud Infra on Friday September 25, 2026 @03:24PM (#49848804)
There are way more than 8 customers.
I have sooo many responsibilities and projects. I often simply don't schedule that type of time.
Re:Ask HN: How do your teams share and distribute agent skills? on Tuesday September 22, 2026 @02:49PM (#49806224)
we have a repo. <org>/<team-name>-skills
Re:Just Bury Your Trash on Tuesday September 01, 2026 @05:50PM (#49528730)
I wasn't disagreeing with the article, simply saying that it still didn't make me feel good.
Re:Just Bury Your Trash on Tuesday September 01, 2026 @12:38PM (#49524379)
Dropped some stuff off at a landfill a few weeks ago. Place was a mess, made me feel like a dirty human.
Ed Zitron will still find something negative to say.
Silly, there will be skills to remove these watermarks.
FOIA request his posts daily?
Re:Ask HN: When did we go from agentic loops to graphs? on Saturday August 01, 2026 @02:27PM (#49137021)
Its a fast moving space. Just getting a firm grasp on one pattern should be a starting point for anyone (past basic prompting). For example I just showed a coworker plan and goal style workflows this week.

Its progress, but IMO its not worth workflow jumping until you've got existing purchase on a working flow.

Re:Mitchellh starts a new company: Superlogical on Wednesday July 29, 2026 @10:12PM (#49105407)
No remote jobs, bummer!
It does take a bit of tokens and some thought to get a workflow that produces output in a way that works for the user. Simple youtube "ai workflows" and you will be inundated with options.
Ya thought about this last weekend. Image a world where youtube knows it you watching and the ad you get is 100% ai generated to you. Like some future world where you're using product x and how fun it would be if you bought it.
Re:The AI Bubble Is No Ordinary Bubble on Tuesday July 21, 2026 @11:03PM (#49001329)
Silly AI videos and just plain odd datacenter zoning/planning aside, the tech is super useful.
With the current state deficit?
Can they build a nuclear power plant or some housing instead?
Re:Ask HN: How do you use LLMs for private discussions? on Sunday July 12, 2026 @07:28PM (#48885931)
Hey sure.

It depends, but usually spin up an h100 on lambda.ai or coreweave. They have capacity and their UIs/APIs are nice. I spin it up for an hour or two, believe it was 6~ dollars an hour.

Once the gpu instance is up, you need to run vllm and a model, ie https://docs.lambda.ai/education/large-language-models/deplo....

Then you can connect your pi.dev, openwebui, etc etc to vllm and interact with it like normal.

Re:Ask HN: How do you use LLMs for private discussions? on Sunday July 12, 2026 @06:17PM (#48885433)
I spin up a gpu instance in a cloud, run my model via vllm, connect to it via an ssh tunnel. done.
SO and friends are in the industry. I will never understand why they simply don't train and accept more Doctors. If they can have 1 person working 80 hours, they can have 2 working 40, or better yet 3 and see more patients and reduce wait times.

Yes there is a cost to training Doctors but you can't tell me its worth the burnout rate?

Older »