<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Neuron AI Blog</title>
    <link>https://noisegg.uk/blog</link>
    <description>Notes from the Neuron AI team: product decisions, engineering deep-dives, and perspectives on multi-model AI.</description>
    <language>en-us</language>
    <atom:link href="https://noisegg.uk/rss.xml" rel="self" type="application/rss+xml" />

    <item>
      <title>Why a multi-model AI workspace is the right shape in 2026</title>
      <link>https://noisegg.uk/blog/multi-model-ai-workspace-2026</link>
      <description>Single-model chat apps are a 2024 pattern. The interesting problem now is letting a user compare four frontier models in the same session.</description>
      <pubDate>Tue, 15 Jul 2026 09:00:00 GMT</pubDate>
      <guid isPermaLink="true">https://noisegg.uk/blog/multi-model-ai-workspace-2026</guid>
    </item>

    <item>
      <title>What model routing actually looks like under the hood</title>
      <link>https://noisegg.uk/blog/what-routing-actually-looks-like</link>
      <description>A walkthrough of how Neuron AI routes a prompt to a model, what we cache, and what we re-send.</description>
      <pubDate>Tue, 08 Jul 2026 09:00:00 GMT</pubDate>
      <guid isPermaLink="true">https://noisegg.uk/blog/what-routing-actually-looks-like</guid>
    </item>

    <item>
      <title>Thinking about LLMs as a workspace, not a vendor</title>
      <link>https://noisegg.uk/blog/thinking-about-llms-as-a-workspace</link>
      <description>What changes when you stop treating AI as a product and start treating it as a tool — and why Neuron AI looks the way it does.</description>
      <pubDate>Wed, 01 Jul 2026 09:00:00 GMT</pubDate>
      <guid isPermaLink="true">https://noisegg.uk/blog/thinking-about-llms-as-a-workspace</guid>
    </item>
  </channel>
</rss>
