<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: samwho</title><link>https://news.ycombinator.com/user?id=samwho</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Wed, 10 Jun 2026 03:14:05 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=samwho" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by samwho in "Project Gutenberg – keeps getting better"]]></title><description><![CDATA[
<p>Looking really good! Great work.</p>
]]></description><pubDate>Sat, 16 May 2026 09:39:57 +0000</pubDate><link>https://news.ycombinator.com/item?id=48158558</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=48158558</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48158558</guid></item><item><title><![CDATA[New comment by samwho in "Quantization from the Ground Up"]]></title><description><![CDATA[
<p>You’re welcome! Thanks so much for the kind words.</p>
]]></description><pubDate>Wed, 25 Mar 2026 20:05:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=47522455</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=47522455</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47522455</guid></item><item><title><![CDATA[New comment by samwho in "Quantization from the Ground Up"]]></title><description><![CDATA[
<p>Thank you!</p>
]]></description><pubDate>Wed, 25 Mar 2026 19:52:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=47522311</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=47522311</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47522311</guid></item><item><title><![CDATA[New comment by samwho in "Quantization from the Ground Up"]]></title><description><![CDATA[
<p>“Zero point” is how I saw it referred to in the literature, so that’s what I went with. I personally prefer to think of it as an offset, but I try to stick with terms folks are likely to see in the wild.</p>
]]></description><pubDate>Wed, 25 Mar 2026 19:52:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=47522308</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=47522308</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47522308</guid></item><item><title><![CDATA[New comment by samwho in "Quantization from the Ground Up"]]></title><description><![CDATA[
<p>Definitely could be, but in the time I spent talking to the 4-bit models in comparison to the 16-bit original it seemed surprisingly capable still. I do recommend benchmarking quantized models at the specific tasks you care about.</p>
]]></description><pubDate>Wed, 25 Mar 2026 17:22:55 +0000</pubDate><link>https://news.ycombinator.com/item?id=47520400</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=47520400</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47520400</guid></item><item><title><![CDATA[New comment by samwho in "Quantization from the Ground Up"]]></title><description><![CDATA[
<p>Thank you! I was really surprised how robust models are to losing information. It seems wrong that they can be compressed so much and still function at all, never mind function quite closely to the original size.<p>Think we're only going to keep seeing more progress in this area on the research side, too.</p>
]]></description><pubDate>Wed, 25 Mar 2026 17:20:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=47520361</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=47520361</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47520361</guid></item><item><title><![CDATA[Quantization from the Ground Up]]></title><description><![CDATA[
<p>Article URL: <a href="https://ngrok.com/blog/quantization">https://ngrok.com/blog/quantization</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47519295">https://news.ycombinator.com/item?id=47519295</a></p>
<p>Points: 351</p>
<p># Comments: 59</p>
]]></description><pubDate>Wed, 25 Mar 2026 16:06:34 +0000</pubDate><link>https://ngrok.com/blog/quantization</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=47519295</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47519295</guid></item><item><title><![CDATA[Don't blunder: better chess puzzles]]></title><description><![CDATA[
<p>Article URL: <a href="https://blunder.clinic">https://blunder.clinic</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47382457">https://news.ycombinator.com/item?id=47382457</a></p>
<p>Points: 3</p>
<p># Comments: 2</p>
]]></description><pubDate>Sat, 14 Mar 2026 23:32:18 +0000</pubDate><link>https://blunder.clinic</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=47382457</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47382457</guid></item><item><title><![CDATA[New comment by samwho in "Temporal: A nine-year journey to fix time in JavaScript"]]></title><description><![CDATA[
<p>Thanks for linking to my silly little quiz in the article! :)</p>
]]></description><pubDate>Wed, 11 Mar 2026 16:04:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=47337387</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=47337387</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47337387</guid></item><item><title><![CDATA[New comment by samwho in "I asked Claude for 37,500 random names, and it can't stop saying Marcus"]]></title><description><![CDATA[
<p>I wrote a tool called llmwalk (<a href="https://github.com/samwho/llmwalk" rel="nofollow">https://github.com/samwho/llmwalk</a>) that’ll deterministically show you the likelihood the top N answers are for a given open model and prompt. No help on frontier models, but maybe helpful if you want to run a similar analysis more quickly on open models!</p>
]]></description><pubDate>Wed, 25 Feb 2026 20:29:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=47157390</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=47157390</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47157390</guid></item><item><title><![CDATA[Calculating the shortest path using just CSS]]></title><description><![CDATA[
<p>Article URL: <a href="https://css-tip.com/graph-theory/">https://css-tip.com/graph-theory/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47045190">https://news.ycombinator.com/item?id=47045190</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 17 Feb 2026 08:55:05 +0000</pubDate><link>https://css-tip.com/graph-theory/</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=47045190</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47045190</guid></item><item><title><![CDATA[What those AI benchmark numbers mean]]></title><description><![CDATA[
<p>Article URL: <a href="https://ngrok.com/blog/ai-benchmarks/">https://ngrok.com/blog/ai-benchmarks/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=46814012">https://news.ycombinator.com/item?id=46814012</a></p>
<p>Points: 5</p>
<p># Comments: 1</p>
]]></description><pubDate>Thu, 29 Jan 2026 18:10:43 +0000</pubDate><link>https://ngrok.com/blog/ai-benchmarks/</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=46814012</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46814012</guid></item><item><title><![CDATA[New comment by samwho in "Graph of All Human Languages"]]></title><description><![CDATA[
<p>My mistake! Misunderstood the underlying dataset. Not sure how to edit it though. Thanks for calling it out.</p>
]]></description><pubDate>Thu, 22 Jan 2026 19:03:52 +0000</pubDate><link>https://news.ycombinator.com/item?id=46723692</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=46723692</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46723692</guid></item><item><title><![CDATA[Graph of All Human Languages]]></title><description><![CDATA[
<p>Article URL: <a href="https://dr.eamer.dev/datavis/poems/language/network.html">https://dr.eamer.dev/datavis/poems/language/network.html</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=46716033">https://news.ycombinator.com/item?id=46716033</a></p>
<p>Points: 4</p>
<p># Comments: 2</p>
]]></description><pubDate>Thu, 22 Jan 2026 06:45:00 +0000</pubDate><link>https://dr.eamer.dev/datavis/poems/language/network.html</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=46716033</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46716033</guid></item><item><title><![CDATA[New comment by samwho in "Ask HN: Share your personal website"]]></title><description><![CDATA[
<p>Oh, you beat me to it! :D</p>
]]></description><pubDate>Wed, 14 Jan 2026 19:06:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=46620987</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=46620987</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46620987</guid></item><item><title><![CDATA[New comment by samwho in "Show HN: Vibe coding a bookshelf with Claude Code"]]></title><description><![CDATA[
<p>I love this, the end result looks so good.<p>Something you don’t really mention in the post is why do this? Do you have an end goal or utility in mind for the book shelf? Is it literally just to track ownership? What do you do with that information?</p>
]]></description><pubDate>Mon, 29 Dec 2025 14:16:26 +0000</pubDate><link>https://news.ycombinator.com/item?id=46420976</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=46420976</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46420976</guid></item><item><title><![CDATA[New comment by samwho in "Ask HN: What are the best engineering blogs with real-world depth?"]]></title><description><![CDATA[
<p>Thank you <3</p>
]]></description><pubDate>Tue, 23 Dec 2025 16:39:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=46366733</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=46366733</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46366733</guid></item><item><title><![CDATA[New comment by samwho in "Prompt caching for cheaper LLM tokens"]]></title><description><![CDATA[
<p>Thank you! <3<p>These are all built with React and CSS animations (or the Web Animations API where I needed it). I’m not very good at React so the code is a real mess. 2 of the components also use threejs for the 3D bits.<p>For the stuff on my personal site, which simonw graciously linked to in another reply, you can see all the code behind my work at <a href="https://github.com/samwho/visualisations" rel="nofollow">https://github.com/samwho/visualisations</a></p>
]]></description><pubDate>Fri, 19 Dec 2025 18:32:18 +0000</pubDate><link>https://news.ycombinator.com/item?id=46329192</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=46329192</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46329192</guid></item><item><title><![CDATA[New comment by samwho in "Prompt caching for cheaper LLM tokens"]]></title><description><![CDATA[
<p>Simon, you’re too kind. Thank you. <3</p>
]]></description><pubDate>Fri, 19 Dec 2025 18:26:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=46329131</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=46329131</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46329131</guid></item><item><title><![CDATA[New comment by samwho in "Prompt caching for cheaper LLM tokens"]]></title><description><![CDATA[
<p>When I was writing this, GPT 5.1 was the latest and it got it right away. It’s the sequence of prime numbers fwiw :)</p>
]]></description><pubDate>Fri, 19 Dec 2025 14:01:42 +0000</pubDate><link>https://news.ycombinator.com/item?id=46325937</link><dc:creator>samwho</dc:creator><comments>https://news.ycombinator.com/item?id=46325937</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46325937</guid></item></channel></rss>