<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Writing on Thoughts of Mike</title>
    <link>https://thoughtsofmike.com/thoughts/</link>
    <description>Recent content in Writing on Thoughts of Mike</description>
    <generator>Hugo</generator>
    <language>en</language>
    <lastBuildDate>Tue, 24 Mar 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://thoughtsofmike.com/thoughts/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Image Encoders</title>
      <link>https://thoughtsofmike.com/thoughts/image-encoder/</link>
      <pubDate>Tue, 24 Mar 2026 00:00:00 +0000</pubDate>
      <guid>https://thoughtsofmike.com/thoughts/image-encoder/</guid>
      <description>&lt;p&gt;I&amp;rsquo;ve been slowly trying to add more of my photography to this site over the last few weeks. I spent a few hours over the weekend adding images to folders and getting things organized. Most of my photography is digital, with a handful of film scans as well. And they take a variety of sizes and aspect ratios. To keep the hosting costs under control, I figured I would crop and resize the images I was adding so they&amp;rsquo;d be a more uniform size. And in the process, I went overboard and ended up making them too small to be useful. And they were all renamed by that point.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Local LLM</title>
      <link>https://thoughtsofmike.com/thoughts/local-llm/</link>
      <pubDate>Sat, 14 Mar 2026 00:00:00 +0000</pubDate>
      <guid>https://thoughtsofmike.com/thoughts/local-llm/</guid>
      <description>&lt;p&gt;I&amp;rsquo;ve been posting a lot about &lt;a href=&#34;https://thoughtsofmike.com/projects/daft-punk-part-1&#34; &gt;my Daft Punk project&lt;/a&gt;, and one of the things that I&amp;rsquo;ve been relying quite heavily on is local large language models for coding assistance. It&amp;rsquo;s one of those things that I thought would be interesting to play with, and ended up actually being kind of useful. I really do think that these kinds of models are going to play a really big role in the future of AI assisted software development and &lt;a href=&#34;https://www.apple.com/newsroom/2026/03/apple-introduces-macbook-pro-with-all-new-m5-pro-and-m5-max/&#34;  class=&#34;external-link&#34; target=&#34;_blank&#34; rel=&#34;noopener&#34;&gt;last week&amp;rsquo;s announcements by Apple about their M5 chips&lt;/a&gt;, combined with &lt;a href=&#34;https://techcrunch.com/2026/03/02/anthropics-claude-reports-widespread-outage/&#34;  class=&#34;external-link&#34; target=&#34;_blank&#34; rel=&#34;noopener&#34;&gt;large scale outages of services like Claude code&lt;/a&gt; make me increasingly convinced that local language models aren&amp;rsquo;t going anywhere. It&amp;rsquo;s really amazing what you can accomplish with an open source model that you&amp;rsquo;ve downloaded for free from the Internet.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Hello, World</title>
      <link>https://thoughtsofmike.com/thoughts/hello-world/</link>
      <pubDate>Tue, 10 Mar 2026 00:00:00 +0000</pubDate>
      <guid>https://thoughtsofmike.com/thoughts/hello-world/</guid>
      <description>&lt;p&gt;This is the first post. But there&amp;rsquo;s plenty more to come.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
