<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Ai on Markus Löning</title><link>https://www.mloning.com/tags/ai/</link><description>Recent content in Ai on Markus Löning</description><generator>Hugo</generator><language>en-us</language><copyright>Copyright © 2025 Markus Löning</copyright><lastBuildDate>Mon, 06 Jul 2026 18:03:47 +0200</lastBuildDate><atom:link href="https://www.mloning.com/tags/ai/index.xml" rel="self" type="application/rss+xml"/><item><title>Agentic engineering and skills</title><link>https://www.mloning.com/posts/agentic-engineering-skills/</link><pubDate>Fri, 05 Jun 2026 14:23:23 +0100</pubDate><guid>https://www.mloning.com/posts/agentic-engineering-skills/</guid><description>&lt;p&gt;AI agents like Claude Code or Codex are non-deterministic tools, which can be modified, tweaked and used in many different ways.
Using them effectively requires mastering them and making them your own to some degree, even when agents and the LLMs they rely on are still evolving.&lt;/p&gt;</description></item><item><title>Working with Claude Code</title><link>https://www.mloning.com/posts/working-with-claude-code/</link><pubDate>Sat, 07 Mar 2026 14:58:15 +0100</pubDate><guid>https://www.mloning.com/posts/working-with-claude-code/</guid><description>&lt;p&gt;Notes and links about Claude Code and agentic engineering.&lt;/p&gt;
&lt;h2 id="how-claude-code-work"&gt;How Claude Code work&lt;/h2&gt;
&lt;p&gt;It&amp;rsquo;s not hard to understand the basic workings of coding agents like Claude Code:&lt;/p&gt;</description></item><item><title>EurIPS 2025</title><link>https://www.mloning.com/posts/eurips-2025/</link><pubDate>Tue, 02 Dec 2025 11:33:55 +0100</pubDate><guid>https://www.mloning.com/posts/eurips-2025/</guid><description>&lt;p&gt;I attended EurIPS 2025 in Copenhagen, the first edition of the European version of NeurIPS.
My notes are mainly about topics related to my current work, that is, time series analysis and biosignals.&lt;/p&gt;</description></item><item><title>Pydata Paris 2025</title><link>https://www.mloning.com/posts/pydata-paris-2025/</link><pubDate>Thu, 02 Oct 2025 11:58:15 +0200</pubDate><guid>https://www.mloning.com/posts/pydata-paris-2025/</guid><description>&lt;ul&gt;
&lt;li&gt;&lt;code&gt;quarto&lt;/code&gt; for publication-ready reports and presentations (see e.g. &lt;a href="https://github.com/paddyroddy/talks"&gt;https://github.com/paddyroddy/talks&lt;/a&gt;)&lt;/li&gt;
&lt;li&gt;an overview of the array API standard by Lucas Colley&lt;/li&gt;
&lt;li&gt;optimal transport theory and Python library (&lt;a href="https://github.com/PythonOT/POT"&gt;https://github.com/PythonOT/POT&lt;/a&gt;)&lt;/li&gt;
&lt;li&gt;free-threaded Python rollout (see &lt;a href="https://labs.quansight.org/blog/free-threaded-python-rollout"&gt;https://labs.quansight.org/blog/free-threaded-python-rollout&lt;/a&gt;, &lt;a href="https://lwn.net/Articles/1025893/"&gt;https://lwn.net/Articles/1025893/&lt;/a&gt;)&lt;/li&gt;
&lt;li&gt;conflict-free replicated data type (CRDT) for lock-free concurrent read/write access, they don&amp;rsquo;t need locks because they accept conflicts but define rules (&amp;ldquo;merge&amp;rdquo; operations) that resolve conflicts (e.g. used collaborative text editors like Google Docs) (see &lt;a href="https://read.thecoder.cafe/p/crdt"&gt;https://read.thecoder.cafe/p/crdt&lt;/a&gt;)&lt;/li&gt;
&lt;li&gt;met the maintainer of tslearn, @charavelg&lt;/li&gt;
&lt;li&gt;Introductory talk on prediction intervals and quantile regression based on scikit-learn, but interface seems incomplete, more complete interfaces exists&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;Talks are available on &lt;a href="https://www.youtube.com/playlist?list=PLGVZCDnMOq0rXq5aghELxbd-QgL8uyg0p"&gt;here&lt;/a&gt;.&lt;/p&gt;</description></item></channel></rss>