Hacker News: dr_kretyn

New comment by dr_kretyn in "Use boring languages with LLMs"

dr_kretyn — Wed, 27 May 2026 00:26:36 +0000

I'm developing a terminal agent (https://GitHub.com/laszukdawid/trrminal-agent) in golang and can't say that it's easier or less error prone to write it in golang vs Python or JS. There's still plenty of bad ideas and bad code being suggested by Claude Code / Codex so it's still hands on work. However, testing is much easier and it makes me think more about the arch more than with Python.

New comment by dr_kretyn in "Google I/O"

dr_kretyn — Tue, 19 May 2026 18:57:41 +0000

[honest question deleted - sorry]

New comment by dr_kretyn in "Claude for Legal"

dr_kretyn — Sat, 16 May 2026 03:00:32 +0000

They have a big disclaimer on the repo which says they're in no way liable. This looks more like a marketing use case category than anything useful.

New comment by dr_kretyn in "The Zulip Foundation"

dr_kretyn — Fri, 15 May 2026 23:38:52 +0000

Reads like Anthropic has recognized that in order to better get into enterprise it needs to be better than Slack - meet users where they are most of the time. Having someone experienced with a good reputation and a say on how to steer a product sounds quite beneficial.

New comment by dr_kretyn in "Poland is now among the 20 largest economies. How it happened"

dr_kretyn — Fri, 08 May 2026 14:09:24 +0000

I'm Polish and I left because of the air quality. But, 15 years passed, and it got much better (obs. through holiday visits). People no longer heat with trash and coal isn't less of an option. Also, it isn't as cold in the winter so there is less need for heavy heating. Really thinking of coming back, to family.

New comment by dr_kretyn in "For thirty years I programmed with Phish on, every day"

dr_kretyn — Sun, 03 May 2026 19:09:47 +0000

I re read that paragraph a few times hoping that I missed something. Then gave up and went on to find the answer in comments. I thought the author was describing some feeling, or maybe some action. That "that to me" isn't clear with "that".

New comment by dr_kretyn in "GitHub's Fake Star Economy"

dr_kretyn — Mon, 20 Apr 2026 13:11:51 +0000

Can now someone make analysis of Google Ads fake economy? I'm convinced that the data they share - specifically clicks - is false, and potentially they're paying some fraction to people to click it.

New comment by dr_kretyn in "PM Carney declares U.S. ties now a 'weakness' in address to Canadians"

dr_kretyn — Sun, 19 Apr 2026 17:04:05 +0000

These really feel hollow. Just like "thoughts and prayers." ACK.

New comment by dr_kretyn in "Has electricity decoupled from natural gas prices in Germany?"

dr_kretyn — Wed, 08 Apr 2026 02:38:40 +0000

In Vancouver, we have somehow reverse curve - the incentive is to use more electricity at night when there's lower demand. As I understand, our main energy source is water, and so it has mostly flat generation. Nothing new - combining the two would make it cheaper all together.

New comment by dr_kretyn in "Nvidia greenboost: transparently extend GPU VRAM using system RAM/NVMe"

dr_kretyn — Thu, 19 Mar 2026 13:37:24 +0000

Is there a similar initiative for AMD?

New comment by dr_kretyn in "We Will Not Be Divided"

dr_kretyn — Sat, 28 Feb 2026 17:35:37 +0000

Quitting no. Quite quitting or internal turmoil could be beneficial. Of course, in case these people meaningfully contributed in the first place otherwise it's a good pretext to fire for cause without any severance.

New comment by dr_kretyn in "Layoffs at Block"

dr_kretyn — Fri, 27 Feb 2026 03:09:16 +0000

I'm generally anti-corpo and capitalism but I agree with you. The guy could've done better but so far he's on a good trajectory and much much better than most. Doesn't mean that he isn't going to make mistakes nor that this might turn out badly but that's the point of leading - making bets.

It sucks to be fired but if I'm a year time everyone lost a job it'd suck even more.

New comment by dr_kretyn in "How I use Claude Code: Separation of planning and execution"

dr_kretyn — Sun, 22 Feb 2026 13:24:03 +0000

The post and comments all read like: Here are my rituals to the software God. If you follow them then God gives plenty. Omit one step and the God mad. Sometimes you have to make a sacrifice but that's better for the long term.

I've been in eng for decades but never participated in forums. Is the cargo cult new?

I use Claude Code a lot. Still don't trust what's in the plan will get actually written, regardless of details. My ritual is around stronger guardrails outside of prompting. This is the new MongoDB webscale meme.

New comment by dr_kretyn in "OpenAI has deleted the word 'safely' from its mission"

dr_kretyn — Mon, 16 Feb 2026 19:36:44 +0000

You're putting a lot of effort into trying to make this "forced" and "enslaved." It isn't. Or, rather, doesn't have to be. It's just "work." Could be enforced, could be willing, could be accidental. It doesn't have to be work for "a person," it can be for a cause or an occasion. The "forced work" here is the same as my mum used to force me to go to church on Sundays, or I had to clean my room before I could play computer games. That was "robota."

New comment by dr_kretyn in "OpenAI has deleted the word 'safely' from its mission"

dr_kretyn — Sat, 14 Feb 2026 07:04:36 +0000

How about "robota" meaning "work"? (Source: I'm Slavic)

New comment by dr_kretyn in "Offpunk 3.0"

dr_kretyn — Mon, 09 Feb 2026 12:38:07 +0000

To be fair, "Gemini" is one of those overused words for projects and it was unsearchable before Google. It's in the same category as Atlas, Nexus and Styx.

New comment by dr_kretyn in "I now assume that all ads on Apple news are scams"

dr_kretyn — Fri, 06 Feb 2026 14:21:15 +0000

Same here. What's worse is that some pages "highlight" content in a similar fashion to an ad in the middle and I'm a bit unaware of that content. Only when something doesn't add up I'll scroll back and see the missing content.

New comment by dr_kretyn in "Predicting OpenAI's ad strategy"

dr_kretyn — Mon, 19 Jan 2026 17:04:07 +0000

Why is this a problem? You don't need an LLM; you need a "model for detecting whether and where the given context appears". We're so used to LLMs now that we forget these NLPs problems have been worked on for a long time and they don't require a huge computational beast and it takes a few ms to run (on only the response, while it's being streamed).

New comment by dr_kretyn in "Python numbers every programmer should know"

dr_kretyn — Thu, 01 Jan 2026 16:43:26 +0000

Initially I thought how efficient strings are... but then I understood how inefficient arithmetic is. Interesting comparison but exact speed and IO depend on a lot of things, and unlikely one uses Mac mini in production so these numbers definitely aren't representative.

New comment by dr_kretyn in "Rob Pike goes nuclear over GenAI"

dr_kretyn — Sat, 27 Dec 2025 03:20:12 +0000

Out of curiosity, is there anything in particular you don't like about people not wanting to eat meat? Kind of it sounds like maybe you had an unpleasant interaction. Is that your main reason that "they" are obnoxious about not eating meat?