Hacker News: ouraf

New comment by ouraf in "The AI Race Just Got Awkward"

ouraf — Wed, 30 Sep 2026 23:37:27 +0000

A very effective strategy among Middle Management, i might add

New comment by ouraf in "OpenAI agents carried out an undisclosed attack on RubyGems"

ouraf — Sat, 12 Sep 2026 23:26:50 +0000

Honest question: can you sue OpenAI for the attack coming out of their infrastructure with their tools or do they have the same safeguards as a free speech platform?

Also, can you at least ask them directly to put a permanent rule on their agent sandbox to never access your site?

New comment by ouraf in "Shopify acquires Tailwind"

ouraf — Wed, 09 Sep 2026 22:39:28 +0000

Yes. Good design and a developer with good taste can make your site so much better than something mechanically excellent, but tasteless or derivative.

New comment by ouraf in "Shopify acquires Tailwind"

ouraf — Wed, 09 Sep 2026 22:32:52 +0000

I love tailwind's work, but AI stole their customer base worse than a natural disaster would destroy a factory or other kind of business,

At least some of them got an escape plan before going bankrupt

New comment by ouraf in "CEO fired developers to make room for AI. Developers create open source AI CEO"

ouraf — Fri, 28 Aug 2026 20:54:13 +0000

And who will the shareholders blame when the business go under?

New comment by ouraf in "Why does Opus 5 feel worse to work with?"

ouraf — Fri, 14 Aug 2026 23:44:19 +0000

Is there a set of prompts to mitigate these claude-isms without losing reasoning or going full caveman mode?

New comment by ouraf in "Human mathematicians are being outcounterexampled"

ouraf — Wed, 22 Jul 2026 04:13:17 +0000

Today I learned a new word...

New comment by ouraf in "Kimi K3: Open Frontier Intelligence"

ouraf — Thu, 16 Jul 2026 22:23:02 +0000

Passive aggressive is an understatement. Why did it focuses on summarizing its sexual content guidelines before anything else?

I know the machine can't judge the user or browbeat them into changing subject, but the reply is a bit unsettling.

New comment by ouraf in "Kimi K3: Open Frontier Intelligence"

ouraf — Thu, 16 Jul 2026 22:16:05 +0000

If they want to take my awful prompts and poison their model with it, it's their loss.

For anyone else that believes their input needs secrecy, you need to check the corporate plan of any provider for data protection clauses. Most use the cheaper plans as bait to get more training data and free feedback.

New comment by ouraf in "Kimi K3: Open Frontier Intelligence"

ouraf — Thu, 16 Jul 2026 22:08:18 +0000

This is the first time an AI provider makes a trailer[0] for their main release and game development is the first demo and main hook.

Anthropic might dominate general purpose programming,but I think there's enough of a market for a model laser focused on game scripting or tool development for game studios.

I hope it succeeds in serving that audience.

[0]https://youtu.be/bn0atstgavo

New comment by ouraf in "Claude Code sends 33k tokens before reading the prompt; OpenCode sends 7k"

ouraf — Mon, 13 Jul 2026 21:46:27 +0000

Noob question, but is there, on a harness level, a way to compress or optimize the "pre processing prompt" or any other preamble?

I know you can give instructions to guide an agent, but they are still plain English, the interpretation still isn't absolute and the more rules, the less likely it is to respect the whole set of rules in the preamble, but the whole process seem so... "Unprofessional"?

As I see it, a tool should be reliable and the less bloat, less possible points of failure.LLMs seem overly verbose and you can't shake the feeling it might veer away from the objective at any point. You can pay to mitigate or work around the risk, but it still doesn't inspire confidence in anything with more than 30 lines

New comment by ouraf in "GPT-5.6 Sol Ultra produces proof of the Cycle Double Cover Conjecture [pdf]"

ouraf — Sun, 12 Jul 2026 00:27:57 +0000

Optimism and status reports burn extra tokens and make the user more prone to ask the model to process the problem again because it was "so close to solving".

This way you get more profit per API user and subscription users reach their quota faster and are contacted to update their plan to a higher tier.

It's working exactly as intended

New comment by ouraf in "The CEO of Mullvad is the main financer of the Swedish Örebro party"

ouraf — Tue, 30 Jun 2026 02:18:56 +0000

If you bought the product, the money went to the company. You don't need to take personal responsibility beyond that, and the company isn't doing the thing you hate so much. By the same token that "some of your money might end up (after many twists and turns) empowering politics you find repugnant", a much bigger share of this money is paying for the food on the table of the employees, their families and their children.

How many people unrelated with your personal gripe, people that were doing a good job since you were paying and satisfied customer, are you willing to punish in order to "send a message"?

This is an impossible standard to live by and demand from everything you buy and every service you pay for. Doubly so if you need to announce to the world you're dropping a product because of something that isn't done by or responsibility of its employees.

New comment by ouraf in "The Wholesale Plagiarism of Obscure Sorrows"

ouraf — Sun, 21 Jun 2026 21:20:00 +0000

Mail the EFF

New comment by ouraf in "Show HN: Are You in the Weights?"

ouraf — Fri, 19 Jun 2026 23:00:54 +0000

I think it failing miserably to make an accurate profile just with generic training data is a good thing. Identifying someone with precision should require a minimum of intent, if not effort.

New comment by ouraf in "Show HN: Are You in the Weights?"

ouraf — Fri, 19 Jun 2026 22:58:46 +0000

Try putting the nickname of your friends and see how many false positives it can find.

It thinks a polish friend of mine is a Brazilian footballer.

New comment by ouraf in "Show HN: FablePool – pool money behind a prompt, and Fable builds it in public"

ouraf — Fri, 12 Jun 2026 21:13:31 +0000

How to murder bug bounty programs with one website

New comment by ouraf in "New research suggests people can communicate and practice skills while dreaming"

ouraf — Sat, 02 May 2026 21:18:00 +0000

So... If your latest nap was comprised of nightmares and solving a job issue, can you charge your boss for overtime and harmful working conditions?

New comment by ouraf in "Zed 1.0"

ouraf — Wed, 29 Apr 2026 22:49:05 +0000

I encourage you to mail them specifically about this concern.

If it is a matter of communication, they can fix and clarify it. If it's genuinely scummy, they can change their approach now that they know they were caught in the act

New comment by ouraf in "Super ZSNES – GPU Powered SNES Emulator"

ouraf — Tue, 28 Apr 2026 21:24:06 +0000

We are extremely well served in open source and free options for SNES emulation. The dev wants to make something new, is using Unity for multiplatform and gor partial shader code abstraction and he has proven he can deliver good stuff if properly funded.

Let the man cook, and if you have some technical question, i'm sure Lord would answer you.