<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: Juvination</title><link>https://news.ycombinator.com/user?id=Juvination</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Fri, 08 May 2026 13:45:46 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=Juvination" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by Juvination in "DeepSeek 4 Flash local inference engine for Metal"]]></title><description><![CDATA[
<p>I love scrolling and reading through this, thinking yeah of course Python is slower than Java, oh wow Rust is pretty on par I wonder what the Java devs did. Then you hit asm and your jaw drops.</p>
]]></description><pubDate>Thu, 07 May 2026 20:46:14 +0000</pubDate><link>https://news.ycombinator.com/item?id=48054716</link><dc:creator>Juvination</dc:creator><comments>https://news.ycombinator.com/item?id=48054716</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48054716</guid></item><item><title><![CDATA[New comment by Juvination in "Nano-vLLM: How a vLLM-style inference engine works"]]></title><description><![CDATA[
<p>Nobody ever said that they were nonexistent before LLMs. When you are investigating and trying to determine if something is AI generated they are the number one indicator.<p>So if you're being accused of just spewing AI, then double down and spew what looks EVEN MORE like AI. What are you even doing?</p>
]]></description><pubDate>Mon, 02 Feb 2026 16:22:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=46857819</link><dc:creator>Juvination</dc:creator><comments>https://news.ycombinator.com/item?id=46857819</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46857819</guid></item><item><title><![CDATA[New comment by Juvination in "Nano-vLLM: How a vLLM-style inference engine works"]]></title><description><![CDATA[
<p>The em dashes really aren't helping their case.</p>
]]></description><pubDate>Mon, 02 Feb 2026 15:51:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=46857365</link><dc:creator>Juvination</dc:creator><comments>https://news.ycombinator.com/item?id=46857365</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46857365</guid></item><item><title><![CDATA[New comment by Juvination in "A Brief History of Ralph"]]></title><description><![CDATA[
<p>I've been working with the Ralphosophy? for iterative behavior in my workflow and it seems pretty promising for cutting out a few manual steps.<p>I still have a manual part which is breaking the design document down into multiple small gh issues after a review but I think that is fine for now.<p>Using codex exec, we start working on a github issue with a supplied design document, creating a PR on completion. Then we perform a review using a review skill madeup which is effectively just a "cite your sources" skill on the review along with Open Questions.<p>Then we iterate through open questions doing a minimum of 3 reviews (somewhat arbitrary but sometimes multiple reviews catch things). Then finally I have I have a step in for checking Sonarcloud, fixing them and pushing the changes. Realistically this step should be broken out into multiple iterations to avoid large context rot.<p>What I miss the most is output, seeing whats going on in either Codex or Claude in real time. I can output the last response but it just gets messy until I make something a bit more formal.</p>
]]></description><pubDate>Mon, 19 Jan 2026 18:25:49 +0000</pubDate><link>https://news.ycombinator.com/item?id=46682636</link><dc:creator>Juvination</dc:creator><comments>https://news.ycombinator.com/item?id=46682636</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46682636</guid></item></channel></rss>