<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>Clay on Bruno Sánchez-Andrade Nuño</title>
    <link>https://brunosan.eu/tags/clay/</link>
    <description>Recent content in Clay on Bruno Sánchez-Andrade Nuño</description>
    <image>
      <title>Bruno Sánchez-Andrade Nuño</title>
      <url>https://brunosan.eu/media/GFC2017s.jpg</url>
      <link>https://brunosan.eu/media/GFC2017s.jpg</link>
    </image>
    <generator>Hugo -- 0.147.0</generator>
    <language>en-us</language>
    <lastBuildDate>Thu, 05 Feb 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://brunosan.eu/tags/clay/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Visions of Earth Intelligence</title>
      <link>https://brunosan.eu/posts/2026-02-05-visions-earth-intelligence/</link>
      <pubDate>Thu, 05 Feb 2026 00:00:00 +0000</pubDate>
      <guid>https://brunosan.eu/posts/2026-02-05-visions-earth-intelligence/</guid>
      <description>From DRIP Earth to the Linear Bet: three paths to AI for Earth — Agentic AI, Embeddings, and Index-based retrieval. An invited talk at BSC AI Factory.</description>
    </item>
    <item>
      <title>The Carbon Footprint of Training Clay v1.5</title>
      <link>https://brunosan.eu/posts/2025-10-22-clay-carbon-footprint/</link>
      <pubDate>Wed, 22 Oct 2025 00:00:00 +0000</pubDate>
      <guid>https://brunosan.eu/posts/2025-10-22-clay-carbon-footprint/</guid>
      <description>We promised to publish Clay&amp;#39;s training emissions a year ago. The answer: ~10–12 tCO₂e — less than our company retreat flights, and a reminder that inference can dwarf training.</description>
    </item>
    <item>
      <title>From Earth Pixels to GeoEmbeddings</title>
      <link>https://brunosan.eu/posts/2025-06-04-from-earth-pixels-geoembeddings/</link>
      <pubDate>Wed, 04 Jun 2025 00:00:00 +0000</pubDate>
      <guid>https://brunosan.eu/posts/2025-06-04-from-earth-pixels-geoembeddings/</guid>
      <description>How foundation models compress Earth observation data into rich mathematical vectors — the technology behind Clay and the future of planetary intelligence.</description>
    </item>
    <item>
      <title>The Quest for &#34;What is Where?&#34;</title>
      <link>https://brunosan.eu/posts/2024-09-10-quest-what-where/</link>
      <pubDate>Tue, 10 Sep 2024 00:00:00 +0000</pubDate>
      <guid>https://brunosan.eu/posts/2024-09-10-quest-what-where/</guid>
      <description>From a keynote at the European Rover Challenge in Krakow: how AI foundational models are transforming Earth observation, and our ability to answer the age-old question of what is where on our planet.</description>
    </item>
  </channel>
</rss>
