<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Data Science on cstenkamp.de</title>
    <link>https://cstenkamp.de/tags/data-science/</link>
    <description>Recent content in Data Science on cstenkamp.de</description>
    <generator>Hugo</generator>
    <language>en</language>
    <managingEditor>cstenkamp@gmail.com (Chris Stenkamp)</managingEditor>
    <webMaster>cstenkamp@gmail.com (Chris Stenkamp)</webMaster>
    <lastBuildDate>Mon, 26 May 2025 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://cstenkamp.de/tags/data-science/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Podcast Episode on LLMs for Software Engineering</title>
      <link>https://cstenkamp.de/portfolio/llms_podcast/</link>
      <pubDate>Mon, 26 May 2025 00:00:00 +0000</pubDate><author>cstenkamp@gmail.com (Chris Stenkamp)</author>
      <guid>https://cstenkamp.de/portfolio/llms_podcast/</guid>
      <description>&lt;p&gt;In this (german-language) episode of Code for Thought, Carina Haupt (DLR) and I talk with host Peter Schmid about large language models in Research Software Engineering: where tools like Copilot genuinely help (routine tasks, tests, small scripts), and where real-world projects expose their limits (builds, requirements, missing project context). I share my current work on &lt;a href=&#34;https://cstenkamp.de/portfolio/paper_repo_link&#34;&gt;matching papers to repositories via embeddings&lt;/a&gt; at the project level rather than just functions - so models carry meaningful context across an entire codebase.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Automatic Meal Recognition</title>
      <link>https://cstenkamp.de/portfolio/visiolab_job/</link>
      <pubDate>Wed, 15 Mar 2023 00:00:00 +0000</pubDate><author>cstenkamp@gmail.com (Chris Stenkamp)</author>
      <guid>https://cstenkamp.de/portfolio/visiolab_job/</guid>
      <description>&lt;video controls preload=&#34;auto&#34; width=&#34;100%&#34;  autoplay muted playsinline class=&#34;html-video&#34;&gt;&#xA;    &lt;source src=&#34;https://cstenkamp.de/portfolio/visiolab_job/visiolab_intro.mp4&#34; type=&#34;video/mp4&#34;&gt;&#xA;  &lt;span&gt;Your browser doesn&#39;t support embedded videos, but don&#39;t worry, you can &lt;a href=&#34;https://cstenkamp.de/portfolio/visiolab_job/visiolab_intro.mp4&#34;&gt;download it&lt;/a&gt; and watch it with your favorite video player!&lt;/span&gt;&#xA;&lt;/video&gt;&#xA;&lt;p&gt;&lt;a href=&#34;https://visolab.io/&#34;&gt;VisoLab&lt;/a&gt; is a startup that creates automated self-checkout registers for canteens, where the meals on the tray are recognized on &lt;em&gt;on edge&lt;/em&gt; via an iPad, allowing the customer to pay for their meals within seconds without requiring an employee.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Gesture Recognition in Large Video Corpora</title>
      <link>https://cstenkamp.de/portfolio/gesture-recognition/</link>
      <pubDate>Wed, 17 Mar 2021 00:00:00 +0000</pubDate><author>cstenkamp@gmail.com (Chris Stenkamp)</author>
      <guid>https://cstenkamp.de/portfolio/gesture-recognition/</guid>
      <description>In this student research job for the &lt;a href=&#34;https://www.redhenlab.org/&#34;&gt;&lt;em&gt;Red Hen Lab&lt;/em&gt;&lt;/a&gt;, I worked on a pipeline that automatically detects hand gestures in large datasets of videos, involving &lt;a href=&#34;https://github.com/CMU-Perceptual-Computing-Lab/openpose&#34;&gt;&lt;em&gt;OpenPose&lt;/em&gt;&lt;/a&gt; for Pose Recognition, Person detection, Person tracking, Scene detection and more, and deployed it on a HPC using Snakemake and Singularity to create a multimodal communication corpus from the Ellen DeGeneres show.</description>
    </item>
    <item>
      <title>Deep Document Splitter</title>
      <link>https://cstenkamp.de/portfolio/inserve_job/</link>
      <pubDate>Fri, 01 Mar 2019 00:00:00 +0000</pubDate><author>cstenkamp@gmail.com (Chris Stenkamp)</author>
      <guid>https://cstenkamp.de/portfolio/inserve_job/</guid>
      <description>An automated pipeline to split a bunch of scanned pages into single documents</description>
    </item>
  </channel>
</rss>
