<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Iryna Dohndorf on foojay.io - Friends of OpenJDK</title><link>https://foojayio.github.io/website/today/author/iryna-dohndorf/</link><description>Articles written by Iryna Dohndorf on foojay.io - Friends of OpenJDK</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Mon, 22 Jun 2026 06:54:00 +0000</lastBuildDate><atom:link href="https://foojayio.github.io/website/today/author/iryna-dohndorf/index.xml" rel="self" type="application/rss+xml"/><item><title>Foojay Podcast #99: Testing the Untestable: LLM Security for Java Developers with Tiberius</title><link>https://foojayio.github.io/website/today/foojay-podcast-99/</link><pubDate>Mon, 22 Jun 2026 06:54:00 +0000</pubDate><guid>https://foojayio.github.io/website/today/foojay-podcast-99/</guid><description>&lt;p&gt;Your AI-powered Java application is live in production. But have you actually tested whether it can be jailbroken or manipulated into leaking data it should never reveal? In this episode, Iryna Dohndorf walks us through Tiberius, an open-source security testing library for LLM applications in Java, and explains why everything you know about unit testing needs a rethink when non-determinism is part of the design.&lt;/p&gt;
&lt;h2 id="h2-0-youtube"&gt;YouTube&lt;/h2&gt;
&lt;div style="position: relative; padding-bottom: 56.25%; height: 0; overflow: hidden;"&gt;
			&lt;iframe allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share; fullscreen" loading="eager" referrerpolicy="strict-origin-when-cross-origin" src="https://www.youtube.com/embed/7bBcTzeevEo?autoplay=0&amp;amp;controls=1&amp;amp;end=0&amp;amp;loop=0&amp;amp;mute=0&amp;amp;start=0" style="position: absolute; top: 0; left: 0; width: 100%; height: 100%; border:0;" title="YouTube video"&gt;&lt;/iframe&gt;
		&lt;/div&gt;

&lt;h2 id="h2-1-podcast-apps"&gt;Podcast Apps&lt;/h2&gt;
&lt;p&gt;You can listen and subscribe to the Foojay Podcast on:&lt;/p&gt;</description></item><item><title>Tiberius: A Security Testing Framework for LLM Applications in Java</title><link>https://foojayio.github.io/website/today/tiberius-a-security-testing-framework-for-llm-applications-in-java/</link><pubDate>Thu, 04 Jun 2026 20:09:09 +0000</pubDate><guid>https://foojayio.github.io/website/today/tiberius-a-security-testing-framework-for-llm-applications-in-java/</guid><description>&lt;p&gt;&lt;em&gt;How do you write a regression test for a system that is non-deterministic by design?&lt;/em&gt;&lt;/p&gt;
&lt;hr&gt;
&lt;ol&gt;
&lt;li&gt;The Problem {#h2-0-1-the-problem}&lt;/li&gt;
&lt;/ol&gt;
&lt;hr&gt;
&lt;p&gt;Large Language Models have moved from research artifacts to production infrastructure. Java applications are embedding them into customer-facing services via Spring Boot, and e.g. LangChain4J &amp;mdash; for document summarization, customer support, healthcare assistance, and financial guidance, to name just a few. The deployment surface is growing faster than the security tooling.&lt;/p&gt;
&lt;p&gt;The vulnerability landscape is empirically well-established. Horlacher, Vifian, and Zagidullina (2026) &lt;strong&gt;[4]&lt;/strong&gt; red-teamed &lt;code&gt;gpt-oss-20b&lt;/code&gt; and found that adversarial techniques achieved alarmingly high Attack Success Rates, while non-adversarial probing exposed pervasive stereotypical defaults &amp;mdash; both consistent across English and Swiss German. Their conclusion: &lt;em&gt;&amp;ldquo;current alignment mechanisms have not fully resolved jailbreaks and inherent bias, posing critical challenges for automated decision-making.&amp;rdquo;&lt;/em&gt;&lt;/p&gt;</description></item><item><title>Foojay Podcast #94: More Than a Blog: How Foojay Connects, Sustains, and Evolves the Java Community</title><link>https://foojayio.github.io/website/today/foojay-podcast-94/</link><pubDate>Mon, 04 May 2026 06:26:00 +0000</pubDate><guid>https://foojayio.github.io/website/today/foojay-podcast-94/</guid><description>&lt;p&gt;Foojay.io, the website for the Friends of OpenJDK, is turning six years old. To celebrate, Frank Delporte headed to JCON in Cologne, Germany, and sat down with twelve members of the Java community to talk about what Foojay means to them, what they learn from each other, and how the community is evolving.&lt;/p&gt;
&lt;p&gt;This episode covers a lot of ground. Sharat Chandar reflects on 25 years in the Java community and why the people are what keep the language alive. Markus Westergren and Iryna Dohndorf both focus on a theme that comes up again and again: developer sustainability. Not just green software, but how you stay healthy, grounded, and relevant when AI is changing everything around you. Markus and his wife researched how developers are reacting to AI, from ignoring it completely to transforming everything they do. Iryna talks about building resilience and robustness as skills, not afterthoughts.&lt;/p&gt;</description></item></channel></rss>