<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: seanjhkwon</title><link>https://news.ycombinator.com/user?id=seanjhkwon</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 19 Nov 2026 20:25:28 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=seanjhkwon" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by seanjhkwon in "Anthropic prices initial public offering"]]></title><description><![CDATA[
<p>Yeah.. the models themselves might not seem to have much moat but the teams that already set their workflows with claude don't easily switch. So I'm thinking the rate at which users maintain their subscriptions and developer habits are the real moat</p>
]]></description><pubDate>Thu, 19 Nov 2026 15:52:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=50561710</link><dc:creator>seanjhkwon</dc:creator><comments>https://news.ycombinator.com/item?id=50561710</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50561710</guid></item><item><title><![CDATA[New comment by seanjhkwon in "Gemini 4 Neon"]]></title><description><![CDATA[
<p>I feel similarly. I always liked gemini/agy but couldn't rely on it because of its long-term task stability. Recent models have gotten much better and I use them even more than claude, but I still think opus 5.5 is on top in terms of reliability.</p>
]]></description><pubDate>Thu, 19 Nov 2026 15:48:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=50561665</link><dc:creator>seanjhkwon</dc:creator><comments>https://news.ycombinator.com/item?id=50561665</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50561665</guid></item><item><title><![CDATA[New comment by seanjhkwon in "Atuin Threads: All your agent sessions on every machine"]]></title><description><![CDATA[
<p>I'd love to give it a try. Can I check it out?</p>
]]></description><pubDate>Wed, 11 Nov 2026 19:34:20 +0000</pubDate><link>https://news.ycombinator.com/item?id=50457461</link><dc:creator>seanjhkwon</dc:creator><comments>https://news.ycombinator.com/item?id=50457461</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50457461</guid></item><item><title><![CDATA[New comment by seanjhkwon in "Tracking Anthropic-discovered vulnerabilities"]]></title><description><![CDATA[
<p>Even if you end up rewriting the suggested fix your own way, getting a clean test case that triggers the bug saves so much time. False positives are what burn maintainer energy, so 9 out of 9 being real bugs is pretty impressive.</p>
]]></description><pubDate>Wed, 11 Nov 2026 12:56:42 +0000</pubDate><link>https://news.ycombinator.com/item?id=50452846</link><dc:creator>seanjhkwon</dc:creator><comments>https://news.ycombinator.com/item?id=50452846</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50452846</guid></item><item><title><![CDATA[New comment by seanjhkwon in "Claude Fable 5.2 and Mythos 5.2"]]></title><description><![CDATA[
<p>At this point, I don't really see much difference in output quality between Opus 5.5 and Fable 5.2. Although, Opus 5.5 seemed to just go on auto pilot mode without give tasks much thoughts unless you give it tasks specifically each time. Fable 5.2 seems to think more on its own.</p>
]]></description><pubDate>Tue, 10 Nov 2026 11:09:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=50438650</link><dc:creator>seanjhkwon</dc:creator><comments>https://news.ycombinator.com/item?id=50438650</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50438650</guid></item></channel></rss>