<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>Eugene Oleinik</title>
    <link>https://evoleinik.com/</link>
    <description>Recent content on Eugene Oleinik</description>
    <image>
      <title>Eugene Oleinik</title>
      <url>https://evoleinik.com/og-image.png</url>
      <link>https://evoleinik.com/og-image.png</link>
    </image>
    <generator>Hugo -- 0.152.2</generator>
    <language>en-us</language>
    <lastBuildDate>Sun, 29 Mar 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://evoleinik.com/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Web 4.0 Is Open Databases</title>
      <link>https://evoleinik.com/posts/web4-open-databases/</link>
      <pubDate>Sun, 29 Mar 2026 00:00:00 +0000</pubDate>
      <guid>https://evoleinik.com/posts/web4-open-databases/</guid>
      <description>I asked my AI agent to plan a family escape from Chiang Mai&amp;rsquo;s burning season. It reverse-engineered a proprietary air quality encoding, scraped flights, checked fuel costs, and compared 15 cities across 10 years of data. This is what Web 4.0 actually looks like: open databases + agents that query them.</description>
    </item>
    <item>
      <title>Agent Experience (AX): 10 Principles for CLI Tools AI Agents Can Actually Use</title>
      <link>https://evoleinik.com/posts/vx-launch/</link>
      <pubDate>Fri, 27 Feb 2026 00:00:00 +0000</pubDate>
      <guid>https://evoleinik.com/posts/vx-launch/</guid>
      <description>We have UX. We have DX. But AI agents are now the primary users of developer tools, and nobody&amp;rsquo;s designing for them. Here are 10 principles from watching real agent behavior through usage telemetry.</description>
    </item>
    <item>
      <title>How We Make Claude Remember: Learnings Over Skills</title>
      <link>https://evoleinik.com/posts/ai-agent-learnings/</link>
      <pubDate>Mon, 02 Feb 2026 00:00:00 +0000</pubDate>
      <guid>https://evoleinik.com/posts/ai-agent-learnings/</guid>
      <description>Skills don&amp;rsquo;t reliably auto-invoke. We built a three-layer system: searchable learnings files, a curation skill, and a post-commit hook that reminds you to document.</description>
    </item>
    <item>
      <title>Serve Markdown to AI Agents (10x Smaller Payloads)</title>
      <link>https://evoleinik.com/posts/markdown-for-agents/</link>
      <pubDate>Mon, 02 Feb 2026 00:00:00 +0000</pubDate>
      <guid>https://evoleinik.com/posts/markdown-for-agents/</guid>
      <description>An entire infrastructure layer (CSS, JS, frameworks) is becoming optional for a growing class of consumers. Here&amp;rsquo;s how to serve markdown via HTTP content negotiation.</description>
    </item>
    <item>
      <title>The AI Data Trap: Why You Can&#39;t Opt Out</title>
      <link>https://evoleinik.com/posts/ai-data-trap/</link>
      <pubDate>Wed, 28 Jan 2026 00:00:00 +0000</pubDate>
      <guid>https://evoleinik.com/posts/ai-data-trap/</guid>
      <description>Your competitor uses Claude or GPT to move faster. If you don&amp;rsquo;t, you fall behind. That&amp;rsquo;s the trap.</description>
    </item>
    <item>
      <title>The Best Agent Architecture Is Already in Your Terminal</title>
      <link>https://evoleinik.com/posts/filesystem-agent-context/</link>
      <pubDate>Mon, 12 Jan 2026 00:00:00 +0000</pubDate>
      <guid>https://evoleinik.com/posts/filesystem-agent-context/</guid>
      <description>&lt;h1 id=&#34;the-best-agent-architecture-is-already-in-your-terminal&#34;&gt;The Best Agent Architecture Is Already in Your Terminal&lt;/h1&gt;
&lt;p&gt;My project&amp;rsquo;s CLAUDE.md file had grown to 55KB—242 learnings crammed into one massive file.&lt;/p&gt;
&lt;p&gt;The problem? Claude prepends this file to every single prompt. A 55KB context file means less room for thinking and acting. Sessions hit context limits faster. Compaction happens sooner.&lt;/p&gt;
&lt;p&gt;I noticed the degradation: sessions became noticeably shorter, context compaction triggered more frequently, and the agent seemed to lose track of longer conversations.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Zero-Friction Database Branching with Neon, Git Hooks, and Claude Code</title>
      <link>https://evoleinik.com/posts/neon-git-branching/</link>
      <pubDate>Wed, 07 Jan 2026 00:00:00 +0000</pubDate>
      <guid>https://evoleinik.com/posts/neon-git-branching/</guid>
      <description>&lt;h1 id=&#34;zero-friction-database-branching-with-neon-git-hooks-and-claude-code&#34;&gt;Zero-Friction Database Branching with Neon, Git Hooks, and Claude Code&lt;/h1&gt;
&lt;p&gt;I&amp;rsquo;ve been refining my Neon database branching setup over the past few months. Here&amp;rsquo;s the current state: fully automated branch lifecycle with zero manual cleanup.&lt;/p&gt;
&lt;h2 id=&#34;the-goal&#34;&gt;The Goal&lt;/h2&gt;
&lt;p&gt;When I &lt;code&gt;git checkout -b feat/x&lt;/code&gt;:&lt;/p&gt;
&lt;ol&gt;
&lt;li&gt;Neon database branch created automatically&lt;/li&gt;
&lt;li&gt;&lt;code&gt;.env.local&lt;/code&gt; updated with the new connection string&lt;/li&gt;
&lt;li&gt;Vercel preview deployment uses the same isolated database&lt;/li&gt;
&lt;/ol&gt;
&lt;p&gt;When I merge and delete the branch:&lt;/p&gt;</description>
    </item>
    <item>
      <title>The Iteration Trap: When AI Makes You a Spectator</title>
      <link>https://evoleinik.com/posts/iteration-trap/</link>
      <pubDate>Sun, 28 Dec 2025 00:00:00 +0000</pubDate>
      <guid>https://evoleinik.com/posts/iteration-trap/</guid>
      <description>The iteration trap is really a clarity trap. You iterate because you don&amp;rsquo;t know what you want.</description>
    </item>
    <item>
      <title>Adding LLM Polish to a Speech-to-Text App</title>
      <link>https://evoleinik.com/posts/adding-llm-polish-to-speech-to-text/</link>
      <pubDate>Mon, 22 Dec 2025 00:00:00 +0000</pubDate>
      <guid>https://evoleinik.com/posts/adding-llm-polish-to-speech-to-text/</guid>
      <description>How I added an optional LLM post-processing step to clean up voice transcriptions, with lessons on prompt engineering and graceful degradation.</description>
    </item>
    <item>
      <title>Building an AI-Powered Changelog GitHub Action</title>
      <link>https://evoleinik.com/posts/ai-changelog-github-action/</link>
      <pubDate>Mon, 22 Dec 2025 00:00:00 +0000</pubDate>
      <guid>https://evoleinik.com/posts/ai-changelog-github-action/</guid>
      <description>How a 87-line inline script became a reusable GitHub Action for AI-generated changelog summaries posted to Slack.</description>
    </item>
    <item>
      <title>CLAUDE.md: Building Persistent Memory for AI Coding Agents</title>
      <link>https://evoleinik.com/posts/claude-md-as-agent-memory/</link>
      <pubDate>Mon, 22 Dec 2025 00:00:00 +0000</pubDate>
      <guid>https://evoleinik.com/posts/claude-md-as-agent-memory/</guid>
      <description>How to use a curated CLAUDE.md file as institutional memory for Claude Code, so your AI agent stops repeating the same mistakes every session.</description>
    </item>
    <item>
      <title>Debugging Random Reboots with Claude Code: A PSU Power Limit Story</title>
      <link>https://evoleinik.com/posts/debugging-random-reboots-with-claude-code/</link>
      <pubDate>Mon, 22 Dec 2025 00:00:00 +0000</pubDate>
      <guid>https://evoleinik.com/posts/debugging-random-reboots-with-claude-code/</guid>
      <description>How I used Claude Code to diagnose random system reboots caused by PSU power limits - and why AI assistants excel at systematic hardware debugging.</description>
    </item>
    <item>
      <title>iTerm2 &#43; tmux -CC: The Remote Development Setup Nobody Talks About</title>
      <link>https://evoleinik.com/posts/iterm2-tmux-control-mode/</link>
      <pubDate>Mon, 22 Dec 2025 00:00:00 +0000</pubDate>
      <guid>https://evoleinik.com/posts/iterm2-tmux-control-mode/</guid>
      <description>How iTerm2&amp;rsquo;s tmux control mode turns remote SSH sessions into native Mac tabs with persistent sessions, proper scrollback, and zero friction.</description>
    </item>
    <item>
      <title>The Loop Changes Everything: Why Embodied AI Breaks Current Alignment Approaches</title>
      <link>https://evoleinik.com/posts/the-loop-changes-everything/</link>
      <pubDate>Mon, 22 Dec 2025 00:00:00 +0000</pubDate>
      <guid>https://evoleinik.com/posts/the-loop-changes-everything/</guid>
      <description>Chat models are stateless and safe by architecture. Embodied robots need persistent loops, memory, and self-models - and that&amp;rsquo;s where alignment gets genuinely hard.</description>
    </item>
    <item>
      <title>Preserve macOS App Permissions Across Rebuilds with Self-Signed Certificates</title>
      <link>https://evoleinik.com/posts/macos-dev-signing-preserve-permissions/</link>
      <pubDate>Sun, 21 Dec 2025 00:00:00 +0000</pubDate>
      <guid>https://evoleinik.com/posts/macos-dev-signing-preserve-permissions/</guid>
      <description>Stop re-granting Accessibility permissions after every rebuild by creating a stable self-signed certificate.</description>
    </item>
    <item>
      <title>Speed Up Syncthing File Sync Discovery (From 11 Seconds to 2)</title>
      <link>https://evoleinik.com/posts/syncthing-faster-sync-discovery/</link>
      <pubDate>Tue, 16 Dec 2025 00:00:00 +0000</pubDate>
      <guid>https://evoleinik.com/posts/syncthing-faster-sync-discovery/</guid>
      <description>Reduce Syncthing&amp;rsquo;s file detection delay from 10+ seconds to under 3 by tuning fsWatcherDelayS on the sending machine.</description>
    </item>
  </channel>
</rss>
