<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: jackxlau</title><link>https://news.ycombinator.com/user?id=jackxlau</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Sun, 14 Jun 2026 22:27:43 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=jackxlau" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by jackxlau in "The MilkV Jupiter 2/SpacemiT K3 (RISC-V vector compute)"]]></title><description><![CDATA[
<p>this is neat</p>
]]></description><pubDate>Sun, 14 Jun 2026 08:24:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=48525297</link><dc:creator>jackxlau</dc:creator><comments>https://news.ycombinator.com/item?id=48525297</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48525297</guid></item><item><title><![CDATA[New comment by jackxlau in "Don't trust large context windows"]]></title><description><![CDATA[
<p>In my own testing I have seen peak performance happen usually within 15-20% of the intended context limit, albeit there are a few optimizations depending on the task quality.</p>
]]></description><pubDate>Sun, 14 Jun 2026 08:17:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=48525267</link><dc:creator>jackxlau</dc:creator><comments>https://news.ycombinator.com/item?id=48525267</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48525267</guid></item><item><title><![CDATA[New comment by jackxlau in "The MilkV Jupiter 2/SpacemiT K3 (RISC-V vector compute)"]]></title><description><![CDATA[
<p>the handshake process to unlock the A100 cores is pretty interesting it reminds me of driving CPU cores like a GPU command queue.</p>
]]></description><pubDate>Thu, 11 Jun 2026 18:20:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=48494359</link><dc:creator>jackxlau</dc:creator><comments>https://news.ycombinator.com/item?id=48494359</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48494359</guid></item><item><title><![CDATA[New comment by jackxlau in "Software Is Made Between Commits"]]></title><description><![CDATA[
<p>I came across the conclusion here since a change sometimes spans several repos, per-repo history optimizes the wrong target.</p>
]]></description><pubDate>Thu, 11 Jun 2026 18:15:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=48494271</link><dc:creator>jackxlau</dc:creator><comments>https://news.ycombinator.com/item?id=48494271</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48494271</guid></item></channel></rss>