<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
  <title>AI Maestro Blog</title>
  <subtitle>Notes on working with AI agents: what we try, what works, what does not, and the evidence behind it.</subtitle>
  <link href="https://getaimaestro.com/blog/feed.xml" rel="self"/>
  <link href="https://getaimaestro.com/blog/"/>
  <id>https://getaimaestro.com/blog/</id>
  <updated>2026-10-28T00:00:00Z</updated>
  <author><name>Juan Peláez</name></author>
  <entry>
    <title>AI Maestro on Google Cloud: a reference deployment and what we tested</title>
    <link href="https://getaimaestro.com/blog/ai-maestro-on-google-cloud.html"/>
    <id>https://getaimaestro.com/blog/ai-maestro-on-google-cloud.html</id>
    <updated>2026-10-28T00:00:00Z</updated>
    <summary>AI agents on one Google Cloud VM: Vertex AI models with no API keys, Google Chat and read-only data access. What we built, what broke, and what we did not test.</summary>
  </entry>
  <entry>
    <title>Pi in AI Maestro: what works, what we fixed, and what it is not</title>
    <link href="https://getaimaestro.com/blog/pi-harness-support.html"/>
    <id>https://getaimaestro.com/blog/pi-harness-support.html</id>
    <updated>2026-10-28T00:00:00Z</updated>
    <summary>Pi agents in AI Maestro: message delivery, chat history, status and messaging tools. What we tested with Gemini on Vertex AI, what we fixed, and its limits.</summary>
  </entry>
  <entry>
    <title>Where should you look? What building a world for AI agents taught us</title>
    <link href="https://getaimaestro.com/blog/human-agent-interaction.html"/>
    <id>https://getaimaestro.com/blog/human-agent-interaction.html</id>
    <updated>2026-10-23T00:00:00Z</updated>
    <summary>What other projects do, the mistakes we made in the World view, what the research says about VR, gamification and agent societies, and how we will know if it works.</summary>
  </entry>
</feed>
