<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>CPU Bytes</title>
    <link>https://cpu-bytes.com</link>
    <description>Clear, citation-ready explainers on distributed systems, LLM architecture, and backend performance.</description>
    <language>en</language>
    <lastBuildDate>Wed, 15 Apr 2026 05:27:36 GMT</lastBuildDate>
    <atom:link href="https://cpu-bytes.com/feed.xml" rel="self" type="application/rss+xml"/>
    <item>
      <title>Urbit Doesn&apos;t Solve Privacy. It Solves Sovereignty.</title>
      <link>https://cpu-bytes.com/blogs/field-notes/urbit-sovereignty-not-privacy</link>
      <guid isPermaLink="true">https://cpu-bytes.com/blogs/field-notes/urbit-sovereignty-not-privacy</guid>
      <description>What Urbit actually does, where the privacy and censorship-resistance arguments break down, and when a user-owned ship genuinely makes sense for a POC.</description>
      <pubDate>Thu, 09 Apr 2026 00:00:00 GMT</pubDate>
      <author>Atish Raina</author>
      <category>Infrastructure</category>
      <category>Decentralization</category>
      <category>Urbit</category>
      <category>Sovereignty</category>
    </item>
    <item>
      <title>Planning in AI Agents</title>
      <link>https://cpu-bytes.com/blogs/ai-agentic-systems/planning-in-ai-agents</link>
      <guid isPermaLink="true">https://cpu-bytes.com/blogs/ai-agentic-systems/planning-in-ai-agents</guid>
      <description>How AI agents decompose complex tasks into structured, executable steps — from static and dynamic planning to integration with tools and memory.</description>
      <pubDate>Thu, 02 Apr 2026 00:00:00 GMT</pubDate>
      <author>Atish Raina</author>
      <category>AI Agents</category>
      <category>Planning</category>
      <category>LLM</category>
    </item>
    <item>
      <title>Memory in AI Agents</title>
      <link>https://cpu-bytes.com/blogs/ai-agentic-systems/memory-in-ai-agents</link>
      <guid isPermaLink="true">https://cpu-bytes.com/blogs/ai-agentic-systems/memory-in-ai-agents</guid>
      <description>How memory works in AI agent systems — from short-term context windows to long-term vector storage, retrieval pipelines, and why memory is fundamentally a retrieval problem.</description>
      <pubDate>Mon, 16 Mar 2026 00:00:00 GMT</pubDate>
      <author>Atish Raina</author>
      <category>AI Agents</category>
      <category>Memory</category>
      <category>LLM</category>
      <category>Vector Databases</category>
    </item>
    <item>
      <title>Tool Layer in AI Agent System</title>
      <link>https://cpu-bytes.com/blogs/ai-agentic-systems/tool-layer-in-ai-agents</link>
      <guid isPermaLink="true">https://cpu-bytes.com/blogs/ai-agentic-systems/tool-layer-in-ai-agents</guid>
      <description>How AI agents interact with external tools, APIs, and services — from function calling schemas to execution patterns, validation, and observability.</description>
      <pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate>
      <author>Atish Raina</author>
      <category>AI Agents</category>
      <category>Tool Layer</category>
      <category>LLM</category>
    </item>
    <item>
      <title>Tokens! What They Actually Are and Why They Drive Your Costs!</title>
      <link>https://cpu-bytes.com/blogs/ai-agentic-systems/tokens</link>
      <guid isPermaLink="true">https://cpu-bytes.com/blogs/ai-agentic-systems/tokens</guid>
      <description>Understanding tokens, tokenization, and how they directly impact your AI agent costs — from BPE to prompt caching strategies.</description>
      <pubDate>Tue, 24 Feb 2026 00:00:00 GMT</pubDate>
      <author>Atish Raina</author>
      <category>AI Agents</category>
      <category>Tokens</category>
      <category>LLM</category>
    </item>
    <item>
      <title>AI Agents as Systems</title>
      <link>https://cpu-bytes.com/blogs/ai-agentic-systems/ai-agents-as-systems</link>
      <guid isPermaLink="true">https://cpu-bytes.com/blogs/ai-agentic-systems/ai-agents-as-systems</guid>
      <description>AI agents aren&apos;t standalone script automation anymore. They&apos;re full-fledged systems, and designing one means you need to think in terms of system design, not a script writer.</description>
      <pubDate>Mon, 23 Feb 2026 00:00:00 GMT</pubDate>
      <author>Atish Raina</author>
      <category>AI Agents</category>
      <category>System Design</category>
      <category>LLM</category>
    </item>
  </channel>
</rss>
