<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
  <title>Commit Issues — Hard to commit. Harder not to.</title>
  <subtitle>An independent editorial blog on data, infrastructure, AI, and privacy.</subtitle>
  <link href="https://commitissues.com/atom.xml" rel="self" type="application/atom+xml"/>
  <link href="https://commitissues.com/" rel="alternate" type="text/html"/>
  <id>https://commitissues.com/</id>
  <updated>2026-04-22T00:00:00.000Z</updated>
  <author><name>Nuno F. Assis</name><uri>https://commitissues.com/about</uri></author>
  <rights>© Nuno F. Assis. All rights reserved.</rights>
  <entry>
    <title><![CDATA[The Case for Sovereign Stacks]]></title>
    <link href="https://commitissues.com/essays/undefined" rel="alternate" type="text/html"/>
    <id>https://commitissues.com/essays/undefined</id>
    <updated>2026-04-22T00:00:00.000Z</updated>
    <summary><![CDATA[European organisations are discovering that cloud dependency is not just a cost problem — it is a legal, strategic, and operational one. Here is what a sovereign stack actually looks like in 2026, and why it is more achievable than the hyperscalers want you to believe.]]></summary>
  </entry>
  <entry>
    <title><![CDATA[On X402 and machine-to-machine payments]]></title>
    <link href="https://commitissues.com/notes/undefined" rel="alternate" type="text/html"/>
    <id>https://commitissues.com/notes/undefined</id>
    <updated>2026-04-15T00:00:00.000Z</updated>
    <summary><![CDATA[X402 is a proposed standard for HTTP-native micropayments. It is also the first protocol I have seen that makes agentic AI billing feel tractable.]]></summary>
  </entry>
  <entry>
    <title><![CDATA[Hetzner CAX: six months of Ampere in production]]></title>
    <link href="https://commitissues.com/notes/undefined" rel="alternate" type="text/html"/>
    <id>https://commitissues.com/notes/undefined</id>
    <updated>2026-04-12T00:00:00.000Z</updated>
    <summary><![CDATA[Six months running production workloads on Hetzner's ARM64 CAX instances. What works, what doesn't, and whether the price-to-performance ratio holds up.]]></summary>
  </entry>
  <entry>
    <title><![CDATA[LangGraph in production: what the docs don't tell you]]></title>
    <link href="https://commitissues.com/notes/undefined" rel="alternate" type="text/html"/>
    <id>https://commitissues.com/notes/undefined</id>
    <updated>2026-03-28T00:00:00.000Z</updated>
    <summary><![CDATA[Three things I wish I had known before deploying LangGraph in production. None of them are in the documentation.]]></summary>
  </entry>
  <entry>
    <title><![CDATA[When Your Training Data Becomes Your Liability]]></title>
    <link href="https://commitissues.com/essays/undefined" rel="alternate" type="text/html"/>
    <id>https://commitissues.com/essays/undefined</id>
    <updated>2026-03-10T00:00:00.000Z</updated>
    <summary><![CDATA[Fine-tuning a model on customer data feels like a competitive advantage. It is — until the model starts leaking what it learned. Here is what privacy-aware ML actually requires, and why most organisations are not doing it.]]></summary>
  </entry>
  <entry>
    <title><![CDATA[A Field Guide to Data Residency in 2026]]></title>
    <link href="https://commitissues.com/essays/undefined" rel="alternate" type="text/html"/>
    <id>https://commitissues.com/essays/undefined</id>
    <updated>2026-02-05T00:00:00.000Z</updated>
    <summary><![CDATA[Data residency requirements have multiplied faster than the tooling to satisfy them. This is a practitioner's guide to what the regulations actually say, what cloud providers actually offer, and how to build systems that are genuinely compliant rather than just defensible in a meeting.]]></summary>
  </entry>
</feed>