<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: undefeated</title><link>https://news.ycombinator.com/user?id=undefeated</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 23 Jul 2026 01:50:58 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=undefeated" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by undefeated in "Personal Computer by Perplexity"]]></title><description><![CDATA[
<p>> always-on, local access to your machine's files, apps, and sessions<p>Uh... how about..., no...? What?<p>> Controllable from any device, anywhere<p>...<p>> There is a kill switch<p>Oh great, sounds like you're confident this is safe then!<p>Someone wake me up, please</p>
]]></description><pubDate>Thu, 12 Mar 2026 11:48:14 +0000</pubDate><link>https://news.ycombinator.com/item?id=47349328</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=47349328</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47349328</guid></item><item><title><![CDATA[New comment by undefeated in "AI agents break rules under everyday pressure"]]></title><description><![CDATA[
<p>Unfortunately, it doesn't look like the EU will actually be able to put any meaningful guardrails on AI, as evidenced by the AI act. I honestly don't know whether it's an off-the-scales level of incompentence or just blatant corruption (I vote both), but in any event, I think AI regulation has already failed, somewhat irreversably.</p>
]]></description><pubDate>Thu, 04 Dec 2025 12:14:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=46146785</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=46146785</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46146785</guid></item><item><title><![CDATA[New comment by undefeated in "AI agents break rules under everyday pressure"]]></title><description><![CDATA[
<p>Uhm... duh?<p>> or cannot control any of the inputs to the llm<p>Seeing as LLMs are non-deterministic, I think even this is not enough of a restriction.</p>
]]></description><pubDate>Thu, 04 Dec 2025 11:32:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=46146410</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=46146410</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46146410</guid></item><item><title><![CDATA[New comment by undefeated in "AI agents break rules under everyday pressure"]]></title><description><![CDATA[
<p>Unlikely, because the free version of ChatGPT isn't really making them any money, so less tokens is actually better — which I assume is why anthropic pushes Haiku models on free users which are not just more quantized but also less wordy.</p>
]]></description><pubDate>Thu, 04 Dec 2025 11:29:55 +0000</pubDate><link>https://news.ycombinator.com/item?id=46146385</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=46146385</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46146385</guid></item><item><title><![CDATA[New comment by undefeated in "AI agents break rules under everyday pressure"]]></title><description><![CDATA[
<p>I think part of the problem is that you need a model to classify the data, which needs to be trained on data that wasn't classified (or a dramatically smaller set of human-classified data), so it's effectively impossible to escape this sort of input bias.<p>Tangentially, I'd be far from the first to point out that these LLMs are now polluting their own training data, which makes filtering simulatenously all the more important and impossible.</p>
]]></description><pubDate>Thu, 04 Dec 2025 11:18:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=46146294</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=46146294</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46146294</guid></item><item><title><![CDATA[New comment by undefeated in "AI agents break rules under everyday pressure"]]></title><description><![CDATA[
<p>I never had a conversation like that — probably because I personally rarely use LLMs to actually generate code for me — but I've somehow subconciously learned to do this myself, especially with clarifying questions.<p>If I find myself needing to ask a clarifying question, I always edit the previous message to ask the next question because the models seem to always force what they said in their clarification into further responses.<p>It's... odd... to find myself conditioned, by the LLM, to the proper manners of conditioning the LLM.</p>
]]></description><pubDate>Thu, 04 Dec 2025 11:10:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=46146245</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=46146245</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46146245</guid></item><item><title><![CDATA[New comment by undefeated in "Why I stopped using JSON for my APIs"]]></title><description><![CDATA[
<p>Complaining about JSON and then proceeding to write your API in Dart is... interesting.</p>
]]></description><pubDate>Tue, 02 Dec 2025 07:32:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=46118629</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=46118629</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46118629</guid></item><item><title><![CDATA[New comment by undefeated in "JavaScript Trademark Update"]]></title><description><![CDATA[
<p>Surely you've never read the job description for a programming position</p>
]]></description><pubDate>Sun, 29 Jun 2025 13:37:27 +0000</pubDate><link>https://news.ycombinator.com/item?id=44413041</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=44413041</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44413041</guid></item><item><title><![CDATA[New comment by undefeated in "Solving `Passport Application` with Haskell"]]></title><description><![CDATA[
<p>It benefits no one to acknowledge that months were spent and the only outcome is millions of dollars wated. Sure, they could try to hold the contractor accountable, but that would put them at risk of being held accountable themselves by their own superiors for hiring that contractor in the first place. So they pretend everything works great, and pay accordingly</p>
]]></description><pubDate>Sun, 29 Jun 2025 13:22:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=44412945</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=44412945</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44412945</guid></item><item><title><![CDATA[New comment by undefeated in "X says it is closing operations in Brazil due to judge's content orders"]]></title><description><![CDATA[
<p>Then again, we'd probably end up with all the same bullshit on Threads</p>
]]></description><pubDate>Sat, 17 Aug 2024 17:14:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=41276044</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=41276044</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41276044</guid></item><item><title><![CDATA[New comment by undefeated in "Show HN: Detecting adblock, without JavaScript, by abusing HTTP 103 responses"]]></title><description><![CDATA[
<p>At least we got some of my all-time favorite GitHub issues tho</p>
]]></description><pubDate>Mon, 25 Mar 2024 19:05:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=39820114</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=39820114</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39820114</guid></item><item><title><![CDATA[New comment by undefeated in "A crowd destroyed a driverless Waymo car in San Francisco"]]></title><description><![CDATA[
<p>News stories are generally considered off-topic <a href="https://news.ycombinator.com/newsguidelines.html">https://news.ycombinator.com/newsguidelines.html</a></p>
]]></description><pubDate>Tue, 13 Feb 2024 23:20:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=39364264</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=39364264</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39364264</guid></item><item><title><![CDATA[New comment by undefeated in "Show HN: Shipp, a deadly simple package manager for your C/C++ projects, in Rust"]]></title><description><![CDATA[
<p>Looks cool! It would probably be useful if there were a way to include dependencies that aren't shipp packages though. Otherwise, since most projects one might depend on right now don't have shipp support, there's a risk of ending up having to use other dependency management solutions alongside shipp, making everything even _more_ complicated rather than simpler, right?</p>
]]></description><pubDate>Thu, 25 Jan 2024 23:12:55 +0000</pubDate><link>https://news.ycombinator.com/item?id=39136769</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=39136769</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39136769</guid></item><item><title><![CDATA[New comment by undefeated in "Japan's earthquake alters coastline, extends it by two football fields"]]></title><description><![CDATA[
<p>Anything but the metric system</p>
]]></description><pubDate>Mon, 08 Jan 2024 23:26:58 +0000</pubDate><link>https://news.ycombinator.com/item?id=38919972</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=38919972</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=38919972</guid></item><item><title><![CDATA[New comment by undefeated in "Ask HN: Do You Miss OOP?"]]></title><description><![CDATA[
<p>Agreed. I recently had to write a package with some quite complex type inference requirements, so I had to resort to classes.<p>I've since had to issue 17 patches (it's been 3 weeks) even though the package has 100% test coverage.</p>
]]></description><pubDate>Tue, 24 May 2022 13:22:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=31491736</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=31491736</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=31491736</guid></item><item><title><![CDATA[New comment by undefeated in "Ask HN: Do you know of a browser based DSL development tool?"]]></title><description><![CDATA[
<p>Was it only a BNF parser generator or did it also help with implementing the language's actual functionality?</p>
]]></description><pubDate>Sun, 22 May 2022 11:35:48 +0000</pubDate><link>https://news.ycombinator.com/item?id=31467032</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=31467032</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=31467032</guid></item><item><title><![CDATA[New comment by undefeated in "Ask HN: Do you know of a browser based DSL development tool?"]]></title><description><![CDATA[
<p>I've never heard of such a thing, but it sounds amazing!<p>Out of curiosity: How exactly did it work (for users)? I assume you'd write a grammar on the one pane, but how did you teach the tool to convert it into the output language? Was that embedded into the grammar?</p>
]]></description><pubDate>Sat, 21 May 2022 15:57:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=31458986</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=31458986</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=31458986</guid></item><item><title><![CDATA[New comment by undefeated in "Ask HN: What do you think of GraphQL?"]]></title><description><![CDATA[
<p>I think it's more like "Company X is big and we have to be like company X if we want to be big too"</p>
]]></description><pubDate>Sat, 21 May 2022 13:49:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=31457889</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=31457889</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=31457889</guid></item><item><title><![CDATA[New comment by undefeated in "Ask HN: Is there a web app to collect emails?"]]></title><description><![CDATA[
<p>Not that I know of, but definitely something I'm gonna put on my 'future project ideas' list.</p>
]]></description><pubDate>Sat, 21 May 2022 13:47:36 +0000</pubDate><link>https://news.ycombinator.com/item?id=31457870</link><dc:creator>undefeated</dc:creator><comments>https://news.ycombinator.com/item?id=31457870</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=31457870</guid></item></channel></rss>