<?xml version="1.0" encoding="UTF-8"?>
<feed xmlns="http://www.w3.org/2005/Atom" xml:lang="en">
    <title>jtatman.github.io</title>
    <subtitle>A chronicle of machine learning and ai-assisted madness</subtitle>
    <link rel="self" type="application/atom+xml" href="https://jtatman.github.io/atom.xml"/>
    <link rel="alternate" type="text/html" href="https://jtatman.github.io"/>
    <generator uri="https://www.getzola.org/">Zola</generator>
    <updated>2026-09-11T00:00:00+00:00</updated>
    <id>https://jtatman.github.io/atom.xml</id>
    <entry xml:lang="en">
        <title>Read a Model&#39;s Context Window Like a Budget, Not Magic</title>
        <published>2026-09-11T00:00:00+00:00</published>
        <updated>2026-09-11T00:00:00+00:00</updated>
        
        <author>
          <name>Unknown</name>
        </author>
        
        <link rel="alternate" type="text/html" href="https://jtatman.github.io/context-window-as-a-budget/"/>
        <id>https://jtatman.github.io/context-window-as-a-budget/</id>
        
        <summary type="html">&lt;p&gt;People talk about a language model&#39;s &quot;memory&quot; like it&#39;s some mysterious, ever-growing
pool of understanding. It isn&#39;t. A context window is a fixed, countable, spendable
resource — closer to a budget than to a brain — and understanding it that way changes
how you should actually work with one.&lt;/p&gt;</summary>
        
    </entry>
    <entry xml:lang="en">
        <title>Why Beads + Mnemoria: Giving an AI Agent Memory Across Sessions</title>
        <published>2026-09-10T00:00:00+00:00</published>
        <updated>2026-09-10T00:00:00+00:00</updated>
        
        <author>
          <name>Unknown</name>
        </author>
        
        <link rel="alternate" type="text/html" href="https://jtatman.github.io/beads-and-mnemoria-agent-memory/"/>
        <id>https://jtatman.github.io/beads-and-mnemoria-agent-memory/</id>
        
        <summary type="html">&lt;p&gt;Every AI coding session has the same quiet problem: when it ends, everything it learned
ends with it. Close the window, and the agent that just spent an hour discovering exactly
why your build was broken knows nothing about it tomorrow. You either re-explain
everything from scratch, or you re-discover the same bug the hard way, again. This
workspace runs two small tools specifically to stop that from happening.&lt;/p&gt;</summary>
        
    </entry>
    <entry xml:lang="en">
        <title>Static Sites vs. WordPress-Style CMSs: An Honest Comparison</title>
        <published>2026-09-09T00:00:00+00:00</published>
        <updated>2026-09-09T00:00:00+00:00</updated>
        
        <author>
          <name>Unknown</name>
        </author>
        
        <link rel="alternate" type="text/html" href="https://jtatman.github.io/static-sites-vs-wordpress/"/>
        <id>https://jtatman.github.io/static-sites-vs-wordpress/</id>
        
        <summary type="html">&lt;p&gt;If the only website tooling you&#39;ve used is something like WordPress, Wix, or Squarespace
— log in, click into an editor, hit publish — the idea of a &quot;static site generator&quot; can
sound like a step backward. No login screen, no dashboard, no database. This site runs
on one (Zola), and the tradeoffs are real in both directions, so here&#39;s the honest
version rather than the sales pitch.&lt;/p&gt;</summary>
        
    </entry>
    <entry xml:lang="en">
        <title>What &quot;AI-Assisted&quot; Actually Means, Day to Day</title>
        <published>2026-09-08T00:00:00+00:00</published>
        <updated>2026-09-08T00:00:00+00:00</updated>
        
        <author>
          <name>Unknown</name>
        </author>
        
        <link rel="alternate" type="text/html" href="https://jtatman.github.io/what-ai-assisted-actually-means/"/>
        <id>https://jtatman.github.io/what-ai-assisted-actually-means/</id>
        
        <summary type="html">&lt;p&gt;&quot;AI-assisted&quot; gets attached to everything right now, and it rarely means anything
specific — it can mean &quot;I asked a chatbot for a paragraph&quot; or &quot;an agent has been running
unsupervised in production for a month.&quot; This entire blog network is genuinely
AI-assisted, in a specific and describable way, so rather than talk about it abstractly,
here&#39;s what that actually looked like on one real piece of work: getting the old
after-dark Zola theme working again.&lt;/p&gt;</summary>
        
    </entry>
    <entry xml:lang="en">
        <title>Regression Trees, Explained for People Who Aren&#39;t Data Scientists</title>
        <published>2026-09-07T00:00:00+00:00</published>
        <updated>2026-09-07T00:00:00+00:00</updated>
        
        <author>
          <name>Unknown</name>
        </author>
        
        <link rel="alternate" type="text/html" href="https://jtatman.github.io/regression-trees-for-the-rest-of-us/"/>
        <id>https://jtatman.github.io/regression-trees-for-the-rest-of-us/</id>
        
        <summary type="html">&lt;p&gt;&quot;Machine learning&quot; gets used as a synonym for &quot;deep learning&quot; so often that people
assume every prediction problem needs a neural network, a GPU, and a few thousand
dollars of cloud compute. Most of the time it doesn&#39;t. One of the most useful tools in
the entire field is something you can explain on a napkin: the regression tree.&lt;/p&gt;</summary>
        
    </entry>
    <entry xml:lang="en">
        <title>Local First, Public When Ready: A Git Setup for AI-Assisted Work</title>
        <published>2026-09-04T00:00:00+00:00</published>
        <updated>2026-09-04T00:00:00+00:00</updated>
        
        <author>
          <name>Unknown</name>
        </author>
        
        <link rel="alternate" type="text/html" href="https://jtatman.github.io/local-first-git-for-ai-assisted-work/"/>
        <id>https://jtatman.github.io/local-first-git-for-ai-assisted-work/</id>
        
        <summary type="html">&lt;p&gt;This entire blog network — the one you&#39;re reading right now — is built by working
alongside an AI agent (Claude Code) on a home server, and it&#39;s the first real post to go
out of that setup. Before writing about anything else, it felt right to write about the
plumbing that makes the rest possible, because it&#39;s a pattern I think more people doing
AI-assisted work should know about: &lt;strong&gt;two git remotes, not one.&lt;/strong&gt;&lt;/p&gt;</summary>
        
    </entry>
</feed>
