<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
    <id>https://arschitectura.com/blog/</id>
    <title>ArsChitectura Blog</title>
    <link href="https://arschitectura.com/feed.xml" rel="self" type="application/atom+xml" />
    <link href="https://arschitectura.com/blog/" rel="alternate" type="text/html" />
    <updated>2026-03-25T00:00:00Z</updated>
    <author>
        <name>Raphaël Géronimi</name>
    </author>
    <entry>
        <id>https://arschitectura.com/blog/machine-learning-in-the-wild/</id>
        <title>Machine Learning in the Wild</title>
        <link href="https://arschitectura.com/blog/machine-learning-in-the-wild/" rel="alternate" type="text/html" />
        <published>2026-03-25T00:00:00Z</published>
        <updated>2026-03-25T00:00:00Z</updated>
        <summary type="text">All successful ML models look the same, but every failed one fails in its own way. After 20 years of deploying models in banking, insurance, e-commerce, and hedge funds, I am starting this blog to share bruised reflections on the vast gap between a theoretical algorithm and a reliable production system - where both success and failure have tangible consequences.</summary>
        <content type="text">All successful ML models look the same, but every failed one fails in its own way. After 20 years of deploying models in banking, insurance, e-commerce, and hedge funds, I am starting this blog to share bruised reflections on the vast gap between a theoretical algorithm and a reliable production system - where both success and failure have tangible consequences.</content>
    </entry>
</feed>
