<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Freelance tech lead og seniorudvikler (Java, AI, .NET, cloud) on emaat</title><link>https://emaat.dk/</link><description>Recent content in Freelance tech lead og seniorudvikler (Java, AI, .NET, cloud) on emaat</description><generator>Hugo</generator><language>da-DK</language><atom:link href="https://emaat.dk/index.xml" rel="self" type="application/rss+xml"/><item><title>Unveal</title><link>https://emaat.dk/projects/unveal/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://emaat.dk/projects/unveal/</guid><description>Fælles lodtrækning som en lille ceremoni, til smagningsklubber og middagsselskaber.</description></item><item><title>Kost &amp; Kurs</title><link>https://emaat.dk/projects/kost-og-kurs/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://emaat.dk/projects/kost-og-kurs/</guid><description>Kalorietælling, hvor et måltid er logget på under 30 sekunder.</description></item><item><title>What is the Style</title><link>https://emaat.dk/projects/what-is-the-style/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://emaat.dk/projects/what-is-the-style/</guid><description>Få dit outfit bedømt mod en stil, helt på telefonen.</description></item><item><title>Marketingagent</title><link>https://emaat.dk/projects/style-match-marketing/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://emaat.dk/projects/style-match-marketing/</guid><description>En AI-agent, der skriver udkast til markedsføring af What is the Style. Intet bliver udgivet, før jeg har godkendt det på Telegram.</description></item><item><title>Journal MCP</title><link>https://emaat.dk/projects/journal-mcp/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://emaat.dk/projects/journal-mcp/</guid><description>Privat MCP-server til journal for tech leads. Claude tænker, serveren husker.</description></item><item><title>Tech briefing</title><link>https://emaat.dk/projects/tech-briefing/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://emaat.dk/projects/tech-briefing/</guid><description>Dagens tech-nyheder som privat podcast, læst højt på ti minutter.</description></item><item><title>The Productivity-Experience Paradox</title><link>https://emaat.dk/articles/the-productivity-experience-paradox/</link><pubDate>Wed, 07 Oct 2026 00:00:00 +0000</pubDate><guid>https://emaat.dk/articles/the-productivity-experience-paradox/</guid><description>&lt;p&gt;&lt;strong&gt;Article:&lt;/strong&gt; &lt;a href="https://annievella.com/posts/the-productivity-experience-paradox/"&gt;The Productivity-Experience Paradox&lt;/a&gt; by Annie Vella&lt;/p&gt;&#10;&lt;h2 id="summary"&gt;Summary&lt;/h2&gt;&#10;&lt;p&gt;Vella presents the second set of findings from her Masters research at the University of Auckland on how AI changes software engineering. The study followed engineers from October 2024 to April 2025 and asked two things at two points in time: how AI coding assistants affected their perceived productivity, and how they affected developer experience, measured with the DevEx framework&amp;rsquo;s three dimensions — feedback loops, cognitive load, and flow state.&lt;/p&gt;</description></item><item><title>Harness Design for Long Running Apps</title><link>https://emaat.dk/articles/harness-design-long-running-apps/</link><pubDate>Sun, 12 Apr 2026 00:00:00 +0000</pubDate><guid>https://emaat.dk/articles/harness-design-long-running-apps/</guid><description>&lt;p&gt;&lt;strong&gt;Article:&lt;/strong&gt; &lt;a href="https://www.anthropic.com/engineering/harness-design-long-running-apps"&gt;Harness design for long-running application development&lt;/a&gt; by Prithvi Rajasekaran&lt;/p&gt;&#10;&lt;h2 id="summary"&gt;Summary&lt;/h2&gt;&#10;&lt;p&gt;Rajasekaran tackles two problems: getting Claude to produce high-quality frontend designs, and getting it to build complete applications without human intervention. Naive long-running setups fail in two ways. Models lose coherence as the context fills — some show &amp;ldquo;context anxiety&amp;rdquo; and wrap up early — and agents asked to judge their own work confidently praise it regardless of quality. Inspired by the generator/discriminator split in GANs, he separates creation from evaluation: a standalone evaluator is still lenient out of the box, but it is far easier to tune an evaluator to be sceptical than to make a generator self-critical.&lt;/p&gt;</description></item><item><title>The Advisor Strategy</title><link>https://emaat.dk/articles/the-advisor-strategy/</link><pubDate>Sun, 12 Apr 2026 00:00:00 +0000</pubDate><guid>https://emaat.dk/articles/the-advisor-strategy/</guid><description>&lt;p&gt;&lt;strong&gt;Article:&lt;/strong&gt; &lt;a href="https://claude.com/blog/the-advisor-strategy"&gt;The advisor strategy: Give agents an intelligence boost&lt;/a&gt; by Anthropic&lt;/p&gt;&#10;&lt;h2 id="summary"&gt;Summary&lt;/h2&gt;&#10;&lt;p&gt;Anthropic introduces the advisor strategy — a cost-effective pattern for building AI agents that pairs Claude Opus as an advisor with a smaller model like Sonnet or Haiku as the executor. The smaller model runs the task end-to-end and calls on Opus only for the hard decisions, giving near Opus-level intelligence at close to Sonnet-level cost.&lt;/p&gt;&#10;&lt;p&gt;The executor calls tools, reads results, and iterates toward a solution on its own. When it hits a decision it can&amp;rsquo;t reasonably solve, it consults Opus, which reads the shared context and returns a plan, a correction, or a stop signal — but never calls tools or produces user-facing output. This inverts the usual sub-agent pattern, where a large orchestrator decomposes work for smaller workers: here the small model drives and escalates, with no decomposition, worker pool, or orchestration logic.&lt;/p&gt;</description></item><item><title>Stop Using Pull Requests</title><link>https://emaat.dk/articles/stop-using-pull-requests/</link><pubDate>Sat, 28 Mar 2026 00:00:00 +0000</pubDate><guid>https://emaat.dk/articles/stop-using-pull-requests/</guid><description>&lt;p&gt;&lt;strong&gt;Article:&lt;/strong&gt; &lt;a href="https://a4al6a.substack.com/p/stop-using-pull-requests"&gt;Stop Using Pull Requests&lt;/a&gt; by Andrea Laforgia&lt;/p&gt;&#10;&lt;h2 id="summary"&gt;Summary&lt;/h2&gt;&#10;&lt;blockquote&gt;&#10;&lt;p&gt;&lt;em&gt;&amp;ldquo;Inspection is too late. The quality, good or bad, is already in the product. Cease dependence on inspection to achieve quality. Eliminate the need for inspection on a mass basis by building quality into the product in the first place.&amp;rdquo;&lt;/em&gt;&lt;/p&gt;&#10;&lt;p&gt;— W. Edwards Deming, &lt;em&gt;Out of the Crisis&lt;/em&gt; (1982)&lt;/p&gt;&#10;&lt;/blockquote&gt;&#10;&lt;hr&gt;&#10;&lt;h3 id="the-uncomfortable-truth"&gt;The Uncomfortable Truth&lt;/h3&gt;&#10;&lt;p&gt;A developer writes code in isolation. Opens a pull request. It sits in a queue. Someone leaves a few comments — mostly about naming and formatting. Clicks approve. Merges.&lt;/p&gt;</description></item></channel></rss>