<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Kimmy — Notes</title><description>Notes to my future self: engineering write-ups, project post-mortems, and stray thoughts.</description><link>https://itskimmy.pages.dev</link><language>en</language><item><title>Taking my work off the hackathon servers</title><link>https://itskimmy.pages.dev/en/blog/hello-from-my-own-site</link><guid isPermaLink="true">https://itskimmy.pages.dev/en/blog/hello-from-my-own-site</guid><description>Almost everything I built used to live on Streamlit or Cloud Run, and then quietly died after the contest. Doing it differently this time.</description><pubDate>Fri, 28 Aug 2026 00:00:00 GMT</pubDate><category>Meta</category></item><item><title>Simulation passed. The board disagreed.</title><link>https://itskimmy.pages.dev/en/blog/post-synthesis-lied-to-me</link><guid isPermaLink="true">https://itskimmy.pages.dev/en/blog/post-synthesis-lied-to-me</guid><description>A FIFO arbiter cost me two weeks. The bug was not in the RTL — it was in my assumptions about what &quot;simulation&quot; means.</description><pubDate>Sat, 20 Jun 2026 00:00:00 GMT</pubDate><category>Verilog</category><category>FPGA</category><category>Debugging</category></item><item><title>Why we extended the ISA instead of bolting on an accelerator</title><link>https://itskimmy.pages.dev/en/blog/isa-extension-vs-accelerator</link><guid isPermaLink="true">https://itskimmy.pages.dev/en/blog/isa-extension-vs-accelerator</guid><description>Post-quantum signing spends almost all its time in Keccak. Given two ways to accelerate it, we took the harder one — because of data movement.</description><pubDate>Sat, 30 May 2026 00:00:00 GMT</pubDate><category>RISC-V</category><category>Post-quantum crypto</category><category>Architecture</category></item></channel></rss>