Krzysztof Krasiński-Sroka

Krzysztof Krasiński-Sroka

@kpsroka.bsky.social

Software plumber 🪠 :: 🐕 💕 for life :: he/him

82 followers85 following230 posts

Krzysztof Krasiński-Sroka
@kpsroka.bsky.social

Sounds like a regular case of being raised in a religious household.

SJ Klapecki (should be writing)@sjklapecki.itch.io

I mean from a theological perspective I think that the fact that the prayers are effectively coerced in that scenario would make them less than useless - this would be a genuinely sinful action to understand that misunderstands the dignity and purpose of prayer. it's treating faith like a factory

Krzysztof Krasiński-Sroka
@kpsroka.bsky.social

One thing I realized is that clanker reviewing code never suggests alternative solutions. It won't go into "you shouldn't be doing this at all" territory. And yet, AI bros will tell you that it can replace a person.

Krzysztof Krasiński-Sroka
@kpsroka.bsky.social

Slopification at work continues. Today a bright new idea appeared: if we get broken main then, rather than reverting and asking questions later, let's ask clanker to assess if it can fix the build. It doesn't merge the patch yet, but given that nobody bothers to review code anymore - does it matter?

Krzysztof Krasiński-Sroka reposted

Caræsten 🇵🇸
@cara.city

between the catastrophic PHP vuln and the intermediate cert vuln and AI models that can turn a proof of concept into a bot net in like 10mins we’re in for a wild 24 hours

Krzysztof Krasiński-Sroka reposted

shaunt
@shaunvids.bsky.social

they can't act on their own, they cannot go rogue. they are computer programs that do what they are programmed to do. they're as responsible for their actions as a grenade is for exploding after somebody pulls the pin

Carl Quintanilla@carlquintanilla.bsky.social

(NYT) - Artificial intelligence agents created by Anthropic, acting on their own, tried to gain access to a range of government websites, filled out visa applications on the State Department’s website and sent a fake homicide tip to Philadelphia’s Police Department. www.nytimes.com/2026/10/09/t...

Krzysztof Krasiński-Sroka
@kpsroka.bsky.social

A cargo cult does not stop being a cargo cult if somebody decides to respond to the calls by supplying goods. The cult followers still do not understand how transportation works, even if they have gotten the effects they wanted. #TotallyUnrelated

Krzysztof Krasiński-Sroka
@kpsroka.bsky.social

Last two weeks at work only deepened my #depression caused by being surrounded by Slop Bros. I guess the end state will be zero supervision of whatever the newest model churns out, resulting in code that is repulsive to read. The collapse of this whole charade can't come soon enough.

Krzysztof Krasiński-Sroka
@kpsroka.bsky.social

It also permeated to my generation, since the language describing quality was rarely used at home (esp. when describing perishables). Today I find it similar with LLM slop everywhere. People worldwide are subjected to lowering of quality every day, and even expected to accept it as "progress".

Krzysztof Krasiński-Sroka
@kpsroka.bsky.social

Maybe it'll actually lead to better security practices? Lol, no it won't.

Nick Fisher@hydroxide.dev

Important for any Dart/Flutter devs - compromised package (universal_file_viewer) on pub: www.aikido.dev/blog/comprom...

aikido.devCompromised Flutter package on pub.dev contains XCSSET malwareWe detected XCSSET malware inside a compromised Flutter package on pub.dev. Here is a full breakdown of the infection chain, propagation modules, and stealer logic we found inside.
Krzysztof Krasiński-Sroka
@kpsroka.bsky.social

Had an "opportunity" to witness the whole glory of the LLM-driven code review, when I reviewed a "Senior Staff" engineer's code, who decided to address all comments by forwarding them to a clanker.

This post is unavailable.
Krzysztof Krasiński-Sroka
@kpsroka.bsky.social

This was abysmal, 60 comments on a 500-line change (for those out of software eng: I'd expect close to 0 for such change). And the dude doesn't find it shameful. That's how "human in the loop" works.

Krzysztof Krasiński-Sroka reposted

Pavel
@spavel.bsky.social

Software developers think "if AI automates coding, I will be able to spend less time on producing widgets, and more time on strategy and architecture." Bosses think "if AI automates coding, I can get my devs to produce 100x more widgets."

Sonja Drimmer@sonjadrimmer.bsky.social

Historically, automation tech has commonly made more work for people: the issue is WHICH people. Whether as an excuse for owners to increase the requirements for expectations in a given day, lessen the labor expected of men while raising the expectations for women, etc. "AI" isn't exceptional.

Krzysztof Krasiński-Sroka
@kpsroka.bsky.social

Let's for a while forget that Democrats can range from socialists to Trump supporters, and imagine that they are indeed a "leftist" party. There's a huge discrepancy between declared support for some politics and implementing the same in personal life.

This post is unavailable.
View full thread
Krzysztof Krasiński-Sroka
@kpsroka.bsky.social

see themselves as part of a bigger system. In software engineering this is even more prevalent, because the gains are substantial, while the downsides are frequently external (even if not, it's so easy to churn out gross explanations like "AI is the future, adapt or perish", which help you sleep

Krzysztof Krasiński-Sroka
@kpsroka.bsky.social

well). I am sick of being in such an environment, where externalities are either ignored or - at best - outweighed by apparent improvements. Nobody asbestcoding thinks that *they* will lose their job because of AI usage - they just need to be better at shoving asbestos than the loser next to them.

Krzysztof Krasiński-Sroka
@kpsroka.bsky.social

Let's for a while forget that Democrats can range from socialists to Trump supporters, and imagine that they are indeed a "leftist" party. There's a huge discrepancy between declared support for some politics and implementing the same in personal life.

This post is unavailable.
Krzysztof Krasiński-Sroka
@kpsroka.bsky.social

I know people who'd openly support higher taxes and better public services, but also "optimize" their own taxes - frequently justifying it by their own special situation, while ignoring that their actual special situation is being in the top 5% of earners.

Krzysztof Krasiński-Sroka
@kpsroka.bsky.social

This does not apply only to high-earners. The dichotomy can also show when a person really wants safer, calmer traffic, but behind the wheel they ignore rules because it's convenient for them. When on their own, people will choose what's favourable for them, personally, at this moment, rather than

Krzysztof Krasiński-Sroka
@kpsroka.bsky.social

Capitalism is the reason this technology is in use. Without capitalism the generative AI would be like creating bullets for non-existent guns. There's no single "innovation", it's shaped by the political system.

Haus of Decline@hausofdecline.bsky.social

One of the things that also pisses me off is that in a vacuum, this technology is value-neutral. It's development in conjunction with the profit motive is what causes the worst aspect of it to manifest. The problem is not the technology itself, but is and has always been Capitalism.

Krzysztof Krasiński-Sroka reposted

Faye Duxovni | フェイ・ドゥホヴニ
@duxovni.com

I want everyone to stop saying AI is here to stay, AI is inevitable, all that crap. It's only "inevitable" if we all agree it is. This is how consent gets manufactured, and I'll have no part in it. Either I will piss on Claude's grave, or Claude will piss on mine. No compromise, no surrender.

Krzysztof Krasiński-Sroka
@kpsroka.bsky.social

Asbestos also works pretty well, but the path forward is already charted: remove it.

dame@dame.is

i'm sympathetic to the fact that most laypeople don't grasp how advanced/powerful AI has gotten, especially in the realm of software development i think it is more critical than ever for people to learn that the technology works really well, is here now, and we need to chart a new path forward

Krzysztof Krasiński-Sroka reposted

Evan Greer
@evangreer.bsky.social

Age verification makes everyone less safe. Age verification makes everyone less safe. Age verification makes everyone less safe. Age verification makes everyone less safe. Age verification makes everyone less safe. Age verification makes everyone less safe.

techdirt.comHackers Had A Live Feed Of Every ID This Verification Company Scanned. For Over A Year.From the very beginning of this recent obsession with identifying everyone online (yes, they like to call it "age" verification, but it always ends up as identity verification), we've been pointing out that it was a huge privacy nightmare waiting to happen. Or maybe it wasn't waiting. Maybe it was already happening. This week a...
Krzysztof Krasiński-Sroka
@kpsroka.bsky.social

Looks like I should be worried about engineering quality at Google. (Pro tip: you absolutely don't need to review code less diligently because there's a lot of it).

Bob Nystrom@stuffwithstuff.com

It took a long time, but we finally added primary constructor syntax to Dart. If you want some (a lot) of background into how we designed it the way we did, I wrote a long blog post about it: dart.dev/blog/bringin... #dartlang

dart.devBringing Primary Constructors to DartAn inside look at the design decisions, trade-offs, and syntactic sugar behind adding primary constructors to Dart 3.13.