<?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>Agentic AI on Vasileios Chroniadis blog</title><link>https://chronvas.github.io/tags/agentic-ai/</link><description>Recent content in Agentic AI on Vasileios Chroniadis blog</description><generator>Hugo</generator><language>en</language><lastBuildDate>Mon, 15 Jun 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://chronvas.github.io/tags/agentic-ai/index.xml" rel="self" type="application/rss+xml"/><item><title>The Power of Neutral Friction- Why the ‘Grill Me’ Skill is a Game-Changer for AI Coding</title><link>https://chronvas.github.io/posts/2026-6-15-the-power-of-neutral-friction-why-the-grill-me-skill-is-a-game-changer-for-ai-coding/</link><pubDate>Mon, 15 Jun 2026 00:00:00 +0000</pubDate><guid>https://chronvas.github.io/posts/2026-6-15-the-power-of-neutral-friction-why-the-grill-me-skill-is-a-game-changer-for-ai-coding/</guid><description>&lt;p&gt;In the rapidly evolving landscape of AI-assisted software engineering, we routinely fall into the “specs-to-code” cycle: a frustrating endless cycle of prompt engineering, trying to perfect instructions rather than engaging with the messy, architectural reality of our thoughts and of our projects.&lt;/p&gt;
&lt;p&gt;The reality is that the specs-to-code approach currently doesn’t work very effectively. LLM’s lack the contextual knowledge, but until the technology evolves, we must find more deliberate ways to work with LLMs.&lt;/p&gt;</description></item><item><title>The Socio-Technical Crossroads - Engineering in an Era of Infinite Output</title><link>https://chronvas.github.io/posts/2026-05-19-the-socio-technical-crossroads--engineering-in-an-era-of-infinite-output/</link><pubDate>Tue, 19 May 2026 00:00:00 +0000</pubDate><guid>https://chronvas.github.io/posts/2026-05-19-the-socio-technical-crossroads--engineering-in-an-era-of-infinite-output/</guid><description>&lt;p&gt;Today I watched an incredible Google I/O presentation by Adam Bender titled &amp;ldquo;Software Engineering at the Tipping Point.&amp;rdquo; It perfectly articulated something I’ve been thinking about a lot: as AI automates the &amp;ldquo;how&amp;rdquo; of writing code, our jobs are rapidly shifting toward the &amp;ldquo;why.&amp;rdquo;&lt;/p&gt;
&lt;p&gt;For decades, as software engineers, we have poured our values, our ethos, and our intentions into products through our code. We used our techne—our technical craft—to express what we valued. We valued mastery over specific languages like Kotlin or framework architectures.&lt;/p&gt;</description></item><item><title>Be careful what you wish for</title><link>https://chronvas.github.io/posts/2026-4-22-be-careful-what-you-wish-for/</link><pubDate>Wed, 22 Apr 2026 00:00:00 +0000</pubDate><guid>https://chronvas.github.io/posts/2026-4-22-be-careful-what-you-wish-for/</guid><description>&lt;p&gt;A tiny story about AI&lt;/p&gt;
&lt;figure&gt;
&lt;a href="https://chronvas.github.io/img/lamp.png"&gt;&lt;img src="https://chronvas.github.io/img/lamp.png"&gt;&lt;/a
&lt;/figure&gt;
&lt;p&gt;Articles praising AI capabilities are everywhere. Limitations? Not so much. That’s what makes small, revealing mishaps like this one fun and worth sharing.&lt;/p&gt;
&lt;hr&gt;
&lt;p&gt;I recently asked an AI coding agent to help me with a small project. The requirement was simple: read a file that sat just outside the project folder, and make some code out of that.&lt;/p&gt;
&lt;p&gt;Because of how the chosen build system works, that file could not be used since it was located just outside of the project folder.&lt;/p&gt;</description></item></channel></rss>