<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>sebszyller.com | Blog</title>
    <link>https://sebszyller.com/blog</link>
    <description>Opinionated takes on machine learning, security &amp; privacy, and policy.</description>
    <language>en-gb</language>
    <lastBuildDate>Mon, 02 Feb 2026 07:10:00 GMT</lastBuildDate>
    <atom:link href="https://sebszyller.com/blog/rss.xml" rel="self" type="application/rss+xml"/>
    <item>
      <title>Best Practices for a New Research Project Repo in 2026</title>
      <link>https://sebszyller.com/blog/2026/researchrepoin2026</link>
      <description>Research code is notoriously sloppy but it doesn&apos;t have to be.</description>
      <pubDate>Mon, 02 Feb 2026 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2026/researchrepoin2026</guid>
    </item>
    <item>
      <title>Looking Back at Learning Rust with LLMs: Works 100% of the Time... 50% of the Time</title>
      <link>https://sebszyller.com/blog/2026/rustllmretrospective</link>
      <description>In 2024, I had a mixed experience of learning Rust with LLMs. Suffice to say, things have changed a lot since then.</description>
      <pubDate>Mon, 19 Jan 2026 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2026/rustllmretrospective</guid>
    </item>
    <item>
      <title>Fairness Is Hardly About DEI</title>
      <link>https://sebszyller.com/blog/2025/fairnessvsdei</link>
      <description>Research gets politicised all the time. For better, or worse. But fairness in machine learning? Seriously?</description>
      <pubDate>Tue, 11 Mar 2025 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2025/fairnessvsdei</guid>
    </item>
    <item>
      <title>DeepSeek Drama -- Model Watermarking to the Rescue</title>
      <link>https://sebszyller.com/blog/2025/deepseekdrama</link>
      <description>DeepSeek is accused of training on ChatGPT outputs. OpenAI has some proof. Do you trust it?</description>
      <pubDate>Sat, 01 Feb 2025 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2025/deepseekdrama</guid>
    </item>
    <item>
      <title>Open Weights Have Nothing to Do with Open Source</title>
      <link>https://sebszyller.com/blog/2024/opensourcellms</link>
      <description>We love open source because we can read and modify the code. We love open weights because we can run models locally. They&apos;re the same thing.. wait, what?!</description>
      <pubDate>Fri, 09 Aug 2024 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2024/opensourcellms</guid>
    </item>
    <item>
      <title>Learning Rust with Large Language Models (Part III): Finding a Needle in a Haystack</title>
      <link>https://sebszyller.com/blog/2024/rustwithllmsneedleinhaystack</link>
      <description>Low mean and high standard deviation is the name of the game when it comes to complex topics and LLMs.</description>
      <pubDate>Tue, 23 Jul 2024 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2024/rustwithllmsneedleinhaystack</guid>
    </item>
    <item>
      <title>Learning Rust with Large Language Models (Part II): an Outdated Manual Written by a Newbie</title>
      <link>https://sebszyller.com/blog/2024/rustwithllmsoutdatedmanual</link>
      <description>Imagine a teacher who has read only the introductory chapters but of all textbooks in existence. Enter learning Rust with LLMs.</description>
      <pubDate>Sun, 03 Mar 2024 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2024/rustwithllmsoutdatedmanual</guid>
    </item>
    <item>
      <title>Food Markets Are a Bad Analogy for Data Marketplaces</title>
      <link>https://sebszyller.com/blog/2024/marketanalogy</link>
      <description> It&apos;s common to refer to data marketplaces using the food market analogy -- it&apos;s easy to understand and unfortunately, provides us with a bad mental model. </description>
      <pubDate>Sun, 21 Jan 2024 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2024/marketanalogy</guid>
    </item>
    <item>
      <title>Learning Rust with Large Language Models (Part I): a Project for 2024</title>
      <link>https://sebszyller.com/blog/2024/rustwithllmsstart</link>
      <description>Bespoke tutoring is one of the promises of large language models. Let&apos;s see if they can help me learn Rust.</description>
      <pubDate>Tue, 02 Jan 2024 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2024/rustwithllmsstart</guid>
    </item>
    <item>
      <title>No One Cares About Large Language Models Anymore</title>
      <link>https://sebszyller.com/blog/2023/llmsnoonecares</link>
      <description> Large language models are just better search. It&apos;s all hype. Everyone works on multimodal and video models now. Right? RIGHT?! </description>
      <pubDate>Sat, 23 Dec 2023 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2023/llmsnoonecares</guid>
    </item>
    <item>
      <title>Content Provenance Needs Critical Mass</title>
      <link>https://sebszyller.com/blog/2023/provenancecriticalmass</link>
      <description> Leica adds provenance support to its latest camera, while OpenAI claims they can&apos;t reliably detect synthetic text. Is content provenance just a pipe dream? </description>
      <pubDate>Wed, 29 Nov 2023 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2023/provenancecriticalmass</guid>
    </item>
    <item>
      <title>Data Marketplaces for Individuals (Still) Don&apos;t Make Sense</title>
      <link>https://sebszyller.com/blog/2023/datamarketplacesrevisited</link>
      <description> Controlling your own data is nice. Decide what you want to share, and charge for it; or don&apos;t share anything at all. But other than data that compromises your privacy, do you have anything worth selling? </description>
      <pubDate>Thu, 02 Nov 2023 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2023/datamarketplacesrevisited</guid>
    </item>
    <item>
      <title>The Security Through Obscurity Moment of Large Language Models... and Money</title>
      <link>https://sebszyller.com/blog/2023/llmsecurityobscurity</link>
      <description>OpenAI is on track to learn all the good security practices the hard way. Probably make a tone of money in the process too.</description>
      <pubDate>Sun, 19 Mar 2023 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2023/llmsecurityobscurity</guid>
    </item>
    <item>
      <title>Finding the Beauty in the Imperfections of Generative Art</title>
      <link>https://sebszyller.com/blog/2022/aiartstyle</link>
      <description>Generative art has taken over the internet by storm. It fails in more than one way but do we really care?</description>
      <pubDate>Mon, 05 Sep 2022 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2022/aiartstyle</guid>
    </item>
    <item>
      <title>Can You Spot a Deepfake?</title>
      <link>https://sebszyller.com/blog/2022/deepfakes</link>
      <description>This looks Photoshopped. This CGI is bad. We either heard or said these more than once. There&apos;s a new cool kid on the block now.</description>
      <pubDate>Wed, 20 Jul 2022 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2022/deepfakes</guid>
    </item>
    <item>
      <title>Kosher Data for Your Ethical Needs</title>
      <link>https://sebszyller.com/blog/2022/kosherdata</link>
      <description>I like my data like I like my coffee -- sourced without exploitation.</description>
      <pubDate>Sat, 21 May 2022 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2022/kosherdata</guid>
    </item>
    <item>
      <title>Why Synthetic Data Is Not Private</title>
      <link>https://sebszyller.com/blog/2022/syntheticdata</link>
      <description>And what&apos;s more, it&apos;s snake oil... unless done right.</description>
      <pubDate>Wed, 04 May 2022 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2022/syntheticdata</guid>
    </item>
    <item>
      <title>Data Marketplaces for Individuals Don&apos;t Make Sense</title>
      <link>https://sebszyller.com/blog/2022/datamarketplaces</link>
      <description>Or at least are challenging. Who doesn&apos;t want to go against big tech. and have agency over selling their own data, right?</description>
      <pubDate>Tue, 15 Mar 2022 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2022/datamarketplaces</guid>
    </item>
    <item>
      <title>On the Difficulty of Cross-disciplinary Communication</title>
      <link>https://sebszyller.com/blog/2021/crossdisciplinarycommunication</link>
      <description>I&apos;m sorry but I don&apos;t speak corporate law. Could you explain it to me in Greek instead?</description>
      <pubDate>Sat, 30 Oct 2021 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2021/crossdisciplinarycommunication</guid>
    </item>
    <item>
      <title>Whose Model Is It Anyway?</title>
      <link>https://sebszyller.com/blog/2021/whosemodelisitanyway</link>
      <description>You gathered the data, sanitised it and trained your fancy model. Your clients love it. Can you prove that it&apos;s yours?</description>
      <pubDate>Wed, 10 Feb 2021 07:10:00 GMT</pubDate>
      <guid>https://sebszyller.com/blog/2021/whosemodelisitanyway</guid>
    </item>
  </channel>
</rss>