<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: incrudible</title><link>https://news.ycombinator.com/user?id=incrudible</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Wed, 30 Sep 2026 20:19:46 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=incrudible" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by incrudible in "Small Models Have Arrived"]]></title><description><![CDATA[
<p>Where are these 180 IQ models? Even Fable needs guidance so it actually produces output I consider on par with what I could have written myself in a domain I am familiar with, and I am certainly not at the frontier in those domains.</p>
]]></description><pubDate>Fri, 28 Aug 2026 09:12:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=49476213</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=49476213</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49476213</guid></item><item><title><![CDATA[New comment by incrudible in "Show HN: The load-bearing vocabulary of Claude"]]></title><description><![CDATA[
<p>Is it actually better though? Can you even measure it? Unguided, Claude will just produce mediocrity which by its own existence becomes almost worthless, and the newer versions become increasingly difficult to guide while also being obnoxious to read. If I did not have to interact with it, it would not be a big deal, but then also I could not get value out of it.</p>
]]></description><pubDate>Thu, 27 Aug 2026 20:48:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=49471037</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=49471037</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49471037</guid></item><item><title><![CDATA[New comment by incrudible in "Vomit: Clean up Claude 5's token output with a separate LLM"]]></title><description><![CDATA[
<p>It is not the jargon per se, it is the style. Other models even in the same family are not as bad, and I still have to babysit it, so the voice matters.</p>
]]></description><pubDate>Thu, 20 Aug 2026 22:53:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=49381377</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=49381377</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49381377</guid></item><item><title><![CDATA[New comment by incrudible in "Vomit: Clean up Claude 5's token output with a separate LLM"]]></title><description><![CDATA[
<p>Opus 5 is literally unbearable to read for me, but more importantly, it is much worse than 4.8 and that one was already annoying. The more they train it on its own output, the more pronounced its idiosyncrasies become.</p>
]]></description><pubDate>Thu, 20 Aug 2026 16:54:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=49377158</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=49377158</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49377158</guid></item><item><title><![CDATA[New comment by incrudible in "Memory prices climb 500% in 12 months"]]></title><description><![CDATA[
<p>Manufacturers <i>know</i> demand will last to the extent that big companies are willing to commit to purchases over a given period of time, but supply today is still constrained by capacity built for the pre LLM era. The 500 billion put in by Samsung/SKHynix will increase modern DRAM by roughly 50% but it wont be fully online until 2030.</p>
]]></description><pubDate>Wed, 19 Aug 2026 05:31:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=49357263</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=49357263</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49357263</guid></item><item><title><![CDATA[New comment by incrudible in "Cerebras CS-4"]]></title><description><![CDATA[
<p>It can definitely create a software stack for you if you hold it right, but the software stack supported by a trillion dollar company with decades of expertise, that <i>also</i> uses AI to improve its stack is <i>probably</i> gonna be better.</p>
]]></description><pubDate>Wed, 19 Aug 2026 05:16:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=49357150</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=49357150</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49357150</guid></item><item><title><![CDATA[New comment by incrudible in "Rust-lang/rust is adopting an LLM policy"]]></title><description><![CDATA[
<p>People are reading this as a slight against single authors, but it is a slight against the belief that anyone deeply understands complex code they wrote even a week ago. It is just a bullshit belief. Yes, you build an intuition about it, but you need to constantly refresh your understanding. That is just how the brain works.<p>Now, if you have program that is highly complex, written by a single author who is not you, hinging on the ongoing goodwill and capability of that single author, that is a liability you do not want to have, operationally. Again, that is just how it is.</p>
]]></description><pubDate>Wed, 05 Aug 2026 19:36:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=49187834</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=49187834</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49187834</guid></item><item><title><![CDATA[New comment by incrudible in "Rust-lang/rust is adopting an LLM policy"]]></title><description><![CDATA[
<p>No codebase worth being deeply invested in is deeply understood by anyone, including the original authors. The eternal maintenance argument is tiresome. If there is a problem in need of fixing, you figure it out, you do not depend on some author being available, and LLMs are exceptionally helpful in this regard as well. Then, if the LLM has trouble figuring out, so will another human. It is a code smell.</p>
]]></description><pubDate>Wed, 05 Aug 2026 07:26:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=49179641</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=49179641</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49179641</guid></item><item><title><![CDATA[New comment by incrudible in "LLMs reward expertise"]]></title><description><![CDATA[
<p>Picking a DVD to watch is a rather inconsequential decision. LLMs already do this sometimes, asking you to pick one of a few options, but without domain expertise you will invariably make worse decisions, but if all the n-th order consequences were first explained to you, that would result in you having built domain expertise, but also erasing most of the speed advantage LLMs give you. Moreover, you will never know about the options that are never presented.<p>Inevitably, this is the new tradeoff to make, above average quality comes from asking for more, and knowing what to ask for comes from expertise.</p>
]]></description><pubDate>Tue, 04 Aug 2026 09:57:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=49166357</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=49166357</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49166357</guid></item><item><title><![CDATA[New comment by incrudible in "Kimi K3 is not cheap"]]></title><description><![CDATA[
<p>You are completely off base with your assumption about what it costs to serve models. Nobody is gonna serve you K3 as cheaply as Anthropic is serving you Opus on any of their plans. Nobody has that kind of money to burn. Kimi does not even sell you a plan right now.</p>
]]></description><pubDate>Wed, 29 Jul 2026 00:39:03 +0000</pubDate><link>https://news.ycombinator.com/item?id=49091949</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=49091949</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49091949</guid></item><item><title><![CDATA[New comment by incrudible in "What is happening to jobs? Separating AI hype from reality"]]></title><description><![CDATA[
<p>That's not how you should use it and LLMs aren't <i>that</i> bad anymore. Either you just accept the quality level of LLM code wholesale, which means accept mediocrity in exchange for volume, or you at least leverage the LLM to improve the process surrounding the production code.</p>
]]></description><pubDate>Sun, 26 Jul 2026 13:50:30 +0000</pubDate><link>https://news.ycombinator.com/item?id=49058247</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=49058247</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49058247</guid></item><item><title><![CDATA[New comment by incrudible in "In defense of not understanding your codebase"]]></title><description><![CDATA[
<p>I read it more as pointing out the inevitable and accepting the implications. You can not call it the result of poor practice when every business beyond a certain size has this problem.<p>Perhaps the problem could be mitigated by keeping the company itself small, but that has nothing to do with programming anymore.</p>
]]></description><pubDate>Thu, 16 Jul 2026 10:08:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=48932501</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=48932501</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48932501</guid></item><item><title><![CDATA[New comment by incrudible in "RISC-V Is Inevitable: State of the Union Keynote Argues"]]></title><description><![CDATA[
<p>The good RISC-V designs are not free though and the free ones are not good. MCUs are not a category of computer to draw lessons from for the broader market.</p>
]]></description><pubDate>Wed, 15 Jul 2026 08:21:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=48917775</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=48917775</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48917775</guid></item><item><title><![CDATA[New comment by incrudible in "Show HN: Ant – A JavaScript runtime and ecosystem"]]></title><description><![CDATA[
<p>In the age of LLMs we can assume that any code was generated by an LLM unless noted otherwise, so I think it is fair to not point out the use of LLMs. However, words like hand built should be reserved for the instances where the code was actually written unassisted. Reading all the code means nothing, there is no such thing as code that was written as if you had written it, unless you have written it.</p>
]]></description><pubDate>Sun, 12 Jul 2026 10:37:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=48879992</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=48879992</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48879992</guid></item><item><title><![CDATA[New comment by incrudible in "Show HN: Ant – A JavaScript runtime and ecosystem"]]></title><description><![CDATA[
<p>It is funny that people now instinctively think of AI when encountering slop, as if the slop that the AI produces was not deeply informed by the slop it trained on. Plus, the quality bar on the web has always been low, masking the true capability of LLMs.</p>
]]></description><pubDate>Sun, 12 Jul 2026 10:30:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=48879976</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=48879976</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48879976</guid></item><item><title><![CDATA[New comment by incrudible in "Grok 4.5"]]></title><description><![CDATA[
<p>This is, again, absurd. According to the definition that a doctor is a person performing witchcraft, do you sincerely believe that a person performing witchcraft is a doctor? Do you sincerely believe that a true statement is true?<p>Let me then rephrase the question: Do you reject the definition that a doctor is anyone who proclaims himself to be a doctor? This is, again, an absurd definition of what a doctor, a woman, or really anything could be.<p>So let me broaden it: Not only do I reject this type of definition for doctors or for women, I reject <i>any</i> such circular definition as absurd. You don't?</p>
]]></description><pubDate>Sat, 11 Jul 2026 18:50:22 +0000</pubDate><link>https://news.ycombinator.com/item?id=48874621</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=48874621</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48874621</guid></item><item><title><![CDATA[New comment by incrudible in "Grok 4.5"]]></title><description><![CDATA[
<p>> You initially claimed that it doesn't matter what "women" means when you ask, "Do you sincerely believe that a person becomes a woman simply by self-declaring it?"<p>No, I said it doesn't matter to <i>to the point I am making</i> what I believe, <i>except</i> that I reject that particular definition, because it is absurd on its face. I may or may not reject any number of other definitions of "woman", it's besides the point.<p>As an analogy (futile, I know): There are many definitions of what a "doctor" could be. For example, I can accept the definition that "a doctor is someone has been licensed to practice medicine". I can't accept the definition that "a doctor is anyone who self-identifies as a doctor".<p>Of course, <i>according to the definition</i> that "a doctor is anyone who self-identifies a doctor", someone who self-identifies as a doctor is a doctor. It's obviously true and obviously absurd.</p>
]]></description><pubDate>Thu, 09 Jul 2026 18:17:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=48850239</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=48850239</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48850239</guid></item><item><title><![CDATA[New comment by incrudible in "Grok 4.5"]]></title><description><![CDATA[
<p>You should join the tautology club.</p>
]]></description><pubDate>Thu, 09 Jul 2026 16:03:05 +0000</pubDate><link>https://news.ycombinator.com/item?id=48848145</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=48848145</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48848145</guid></item><item><title><![CDATA[New comment by incrudible in "Grok 4.5"]]></title><description><![CDATA[
<p>> You're making claims about these laws<p>No, I'm making the claim that <i>any such law</i> conflicts with the position that "the state should stay out of it", obviously. And since you are not even the person supposedly holding that position, I don't even know where you're trying get here.<p>> You asked me this: "So you reject laws that require people to affirm my gender based on nothing but my self identification?"<p>I did not. I asked CamperBob2 that, who presumably holds the position that "the government should stay out of it". So unless that is your alt, I think you're missing the point and barking up the wrong tree.<p>> As I have pointed out above, the word "woman" has different definitions. So if you ask me whether I believe that "a person becomes a woman simply by self-declaring it", then the answer is yes for some definitions of the term, and no for others, like I already explained to you.<p>"A woman is any person who self-identifies as a woman" is <i>the definition</i> that started this whole subthread, and a self-referential one at that. The corollary of that definition is that anyone gets to be a woman by simply declaring it. So, according to the definition that a woman is anyone who says they're a woman, anyone who says they're a woman is a woman. That's a tautology, so it's obviously true. Check mate, athiests.</p>
]]></description><pubDate>Thu, 09 Jul 2026 15:02:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=48847103</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=48847103</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48847103</guid></item><item><title><![CDATA[New comment by incrudible in "Grok 4.5"]]></title><description><![CDATA[
<p>> You're still not clarifying what you're actually saying, so I would level the same accusation against you.<p>What clarification do you need? You can pick <i>any</i> law related to gender identity, and that by definition violates the principle of having the "government stay out of it". And I'm not trying to debate the merits of any such law, I'm trying to get to admit the person in question that they don't actually hold the belief that the government "should stay out of it".<p>> It would be helpful if you could be more specific.<p>No, because it actually doesn't matter the point, as outlined above. Again, I'm not interested in a debate on the merits of any such law.<p>> So if your tactic is to "put me on the spot", I would encourage you to apply this to yourself and genuinely consider what you actually believe, and why you believe it.<p>You expend a lot of words dodging a simple question: Do you <i>sincerely</i> believe that a person becomes a woman simply by self-declaring it? I don't believe this. I don't think you believe this.<p>I'm not asking you to define what makes a woman. It doesn't matter to the point what you or I believe makes a woman, except the distinction that it is (or isn't) as simple as declaring it.</p>
]]></description><pubDate>Thu, 09 Jul 2026 10:48:56 +0000</pubDate><link>https://news.ycombinator.com/item?id=48843792</link><dc:creator>incrudible</dc:creator><comments>https://news.ycombinator.com/item?id=48843792</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48843792</guid></item></channel></rss>