<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/"><channel><title>Interlusion Blog</title><description>Gedanken zu AI-nativer Entwicklung, Agenten-Workflows, Design und Engineering.</description><link>https://interlusion.dev/blog/</link><language>de</language><atom:link href="https://interlusion.dev/blog/rss.xml" rel="self" type="application/rss+xml"/><item><title>Denken in einer geliehenen Sprache</title><link>https://interlusion.dev/blog/denken-in-einer-geliehenen-sprache/</link><guid isPermaLink="true">https://interlusion.dev/blog/denken-in-einer-geliehenen-sprache/</guid><description>KI-Modelle denken in einer Sprache, die sie von uns geliehen haben. Das Training schreibt sie leise um. Ein Essay über Komposita, Telegramme und Übergaben.</description><pubDate>Wed, 28 Oct 2026 00:00:00 GMT</pubDate><category>philosophy</category><category>research</category></item><item><title>Doku macht wieder Spaß</title><link>https://interlusion.dev/blog/doku-macht-wieder-spass/</link><guid isPermaLink="true">https://interlusion.dev/blog/doku-macht-wieder-spass/</guid><description>Interne Doku veraltete, sobald sie fertig war. Jetzt schreibt und pflegt KI sie und nutzt sie als Kontext. Aus der ältesten Pflicht wird ein Kreislauf.</description><pubDate>Wed, 28 Oct 2026 00:00:00 GMT</pubDate><category>ai-native</category><category>engineering</category><category>opinion</category></item><item><title>LLMs als stochastische Papageien abzutun, ist keine Analyse</title><link>https://interlusion.dev/blog/llms-als-stochastische-papageien-abzutun-ist-keine-analyse/</link><guid isPermaLink="true">https://interlusion.dev/blog/llms-als-stochastische-papageien-abzutun-ist-keine-analyse/</guid><description>Die Papageien-Metapher verwechselt Mechanismus mit Fähigkeit: Statistik beschreibt, wie ein System arbeitet, nicht, was es kann. Die Debatte verdient mehr.</description><pubDate>Tue, 27 Oct 2026 00:00:00 GMT</pubDate><category>opinion</category><category>philosophy</category></item><item><title>Warum KI-Modelle auf Schmierzetteln denken</title><link>https://interlusion.dev/blog/warum-ki-modelle-auf-schmierzetteln-denken/</link><guid isPermaLink="true">https://interlusion.dev/blog/warum-ki-modelle-auf-schmierzetteln-denken/</guid><description>Reasoning-Modelle schreiben sich Notizen, bevor sie antworten. Warum sie diesen Schmierzettel brauchen, warum er unleserlich wird und warum das zählt.</description><pubDate>Tue, 27 Oct 2026 00:00:00 GMT</pubDate><category>research</category><category>agents</category></item><item><title>Context7 und die Infrastrukturschicht, die Ihren KI-Agenten fehlt</title><link>https://interlusion.dev/blog/context7-und-die-infrastrukturschicht-die-ihren-ki-agenten-fehlt/</link><guid isPermaLink="true">https://interlusion.dev/blog/context7-und-die-infrastrukturschicht-die-ihren-ki-agenten-fehlt/</guid><description>Context7 verhindert, dass Agenten APIs erfinden. Eine Lücke in der Supply Chain zeigt, worum es eigentlich geht: Wem trauen Sie zu, den Kontext zu füllen?</description><pubDate>Mon, 26 Oct 2026 00:00:00 GMT</pubDate><category>agents</category><category>engineering</category><category>tools</category></item><item><title>Ihre KI-Agenten scheitern, weil Sie sie nie eingearbeitet haben</title><link>https://interlusion.dev/blog/ihre-ki-agenten-scheitern-weil-sie-sie-nie-eingearbeitet-haben/</link><guid isPermaLink="true">https://interlusion.dev/blog/ihre-ki-agenten-scheitern-weil-sie-sie-nie-eingearbeitet-haben/</guid><description>Prompt Engineering war nur das Vorprogramm. Ob Agenten abdriften oder besser werden, entscheidet Context Engineering: welche Information wann ankommt.</description><pubDate>Sun, 25 Oct 2026 00:00:00 GMT</pubDate><category>agents</category><category>engineering</category><category>opinion</category></item><item><title>ORMs sind Mist. Welche Alternativen gibt es?</title><link>https://interlusion.dev/blog/orms-sind-mist-welche-alternativen-gibt-es/</link><guid isPermaLink="true">https://interlusion.dev/blog/orms-sind-mist-welche-alternativen-gibt-es/</guid><description>Objekte und Tabellen passen nicht zusammen. ORM-Abfragesprachen schaffen neue Komplexität. Und KI-Agenten schreiben besseres SQL als ORM-Code.</description><pubDate>Sat, 24 Oct 2026 00:00:00 GMT</pubDate><category>engineering</category><category>opinion</category></item><item><title>Was AI-native wirklich bedeutet und warum es zählt</title><link>https://interlusion.dev/blog/was-ai-native-wirklich-bedeutet/</link><guid isPermaLink="true">https://interlusion.dev/blog/was-ai-native-wirklich-bedeutet/</guid><description>AI-native ist kein Marketing-Etikett, sondern eine andere Art, Software zu bauen, zu betreiben und auszuliefern. So sieht das in der Praxis aus.</description><pubDate>Fri, 23 Oct 2026 00:00:00 GMT</pubDate><category>ai-native</category><category>opinion</category></item><item><title>Die beste Sprache für KI-Agenten ist vermutlich nicht Ihre Lieblingssprache</title><link>https://interlusion.dev/blog/die-beste-sprache-fuer-ki-agenten-ist-nicht-ihre-lieblingssprache/</link><guid isPermaLink="true">https://interlusion.dev/blog/die-beste-sprache-fuer-ki-agenten-ist-nicht-ihre-lieblingssprache/</guid><description>Wenn KI-Agenten API-Backends schreiben und pflegen, zählen andere Kriterien: Vorhersagbarkeit und schnelles Feedback schlagen Eleganz. Das spricht für Go.</description><pubDate>Wed, 21 Oct 2026 00:00:00 GMT</pubDate><category>agents</category><category>engineering</category><category>opinion</category><category>go</category></item><item><title>Die haben mich geriggt</title><link>https://interlusion.dev/blog/die-haben-mich-geriggt/</link><guid isPermaLink="true">https://interlusion.dev/blog/die-haben-mich-geriggt/</guid><description>Darbit, der Hase im Interlusion-Logo, übers Animiertwerden: biegsame Ohren, maskierte Augen, ein Zylinder, der sich dreht, ohne sich zu bewegen.</description><pubDate>Fri, 16 Oct 2026 00:00:00 GMT</pubDate><category>ui</category><category>engineering</category></item><item><title>Stitch: vibe design and the DESIGN.md pipeline</title><link>https://interlusion.dev/en/blog/google-stitch-vibe-design-and-the-design-md-pipeline/</link><guid isPermaLink="true">https://interlusion.dev/en/blog/google-stitch-vibe-design-and-the-design-md-pipeline/</guid><description>Google&apos;s AI design tool generates high-fidelity UI from prompts, sketches, and voice. The interesting part is DESIGN.md and its MCP bridge to your code.</description><pubDate>Mon, 30 Mar 2026 00:00:00 GMT</pubDate><dc:language>en</dc:language><category>ui</category><category>tools</category><category>ai-native</category></item><item><title>Paper: the code-native design canvas</title><link>https://interlusion.dev/en/blog/paper-the-code-native-design-canvas/</link><guid isPermaLink="true">https://interlusion.dev/en/blog/paper-the-code-native-design-canvas/</guid><description>Paper eliminates the design-to-code translation by making the canvas itself HTML and CSS. What you design is what ships. The question is whether that&apos;s always what you want.</description><pubDate>Sun, 29 Mar 2026 00:00:00 GMT</pubDate><dc:language>en</dc:language><category>ui</category><category>tools</category><category>ai-native</category></item><item><title>Pencil: design that lives where the code lives</title><link>https://interlusion.dev/en/blog/pencil-design-that-lives-where-the-code-lives/</link><guid isPermaLink="true">https://interlusion.dev/en/blog/pencil-design-that-lives-where-the-code-lives/</guid><description>Pencil puts a Figma-like canvas inside your IDE with design files in Git and AI agents reading specs through MCP. The design never leaves the codebase.</description><pubDate>Sat, 28 Mar 2026 00:00:00 GMT</pubDate><dc:language>en</dc:language><category>ui</category><category>tools</category><category>ai-native</category></item><item><title>The design handoff is dying</title><link>https://interlusion.dev/en/blog/the-design-handoff-is-dying/</link><guid isPermaLink="true">https://interlusion.dev/en/blog/the-design-handoff-is-dying/</guid><description>Three new tools close the gap between design and code from different directions. The deeper shift: agents need a shared design vocabulary to work at all.</description><pubDate>Fri, 27 Mar 2026 00:00:00 GMT</pubDate><dc:language>en</dc:language><category>ui</category><category>opinion</category><category>ai-native</category></item><item><title>Astro: the framework that ships less</title><link>https://interlusion.dev/en/blog/astro-the-framework-that-ships-less/</link><guid isPermaLink="true">https://interlusion.dev/en/blog/astro-the-framework-that-ships-less/</guid><description>Why Interlusion builds its marketing site with Astro, and what a zero-JavaScript-by-default architecture means for performance, simplicity, and AI-native development.</description><pubDate>Thu, 26 Mar 2026 00:00:00 GMT</pubDate><dc:language>en</dc:language><category>engineering</category><category>tools</category><category>ui</category></item><item><title>Building a design token system with Tailwind v4</title><link>https://interlusion.dev/en/blog/building-a-design-token-system-with-tailwind-v4/</link><guid isPermaLink="true">https://interlusion.dev/en/blog/building-a-design-token-system-with-tailwind-v4/</guid><description>A walkthrough of a three-layer design token architecture with Tailwind CSS v4&apos;s native @theme directive. No config files, no build tools, just CSS.</description><pubDate>Tue, 24 Mar 2026 00:00:00 GMT</pubDate><dc:language>en</dc:language><category>ui</category><category>engineering</category><category>tools</category></item><item><title>OKLCH: the end of unpredictable colour</title><link>https://interlusion.dev/en/blog/oklch-the-end-of-unpredictable-color/</link><guid isPermaLink="true">https://interlusion.dev/en/blog/oklch-the-end-of-unpredictable-color/</guid><description>HSL promised human-readable colour and lied about lightness. OKLCH delivers: perceptually uniform colour for P3 displays, design tokens, and AI agents.</description><pubDate>Tue, 24 Mar 2026 00:00:00 GMT</pubDate><dc:language>en</dc:language><category>ui</category><category>engineering</category></item><item><title>Design tokens: the API between design and code</title><link>https://interlusion.dev/en/blog/design-tokens-the-api-between-design-and-code/</link><guid isPermaLink="true">https://interlusion.dev/en/blog/design-tokens-the-api-between-design-and-code/</guid><description>Design systems fail when design and code drift apart. Design tokens fix the interface: one source of truth that both sides can read, write, and ship from.</description><pubDate>Mon, 23 Mar 2026 00:00:00 GMT</pubDate><dc:language>en</dc:language><category>ui</category><category>engineering</category><category>opinion</category></item><item><title>PageIndex and the case against chunking</title><link>https://interlusion.dev/en/blog/pageindex-and-the-case-against-chunking/</link><guid isPermaLink="true">https://interlusion.dev/en/blog/pageindex-and-the-case-against-chunking/</guid><description>VectifyAI&apos;s PageIndex swaps vector search for tree reasoning and scores 98.7% on FinanceBench. The result is real. The question is what it actually proves.</description><pubDate>Fri, 20 Mar 2026 00:00:00 GMT</pubDate><dc:language>en</dc:language><category>research</category><category>engineering</category><category>rag</category></item><item><title>Most RAG fails at the R, not the G</title><link>https://interlusion.dev/en/blog/most-rag-fails-at-the-r-not-the-g/</link><guid isPermaLink="true">https://interlusion.dev/en/blog/most-rag-fails-at-the-r-not-the-g/</guid><description>Everyone obsesses over models and prompts. The real bottleneck in RAG is retrieval, and fixing it takes engineering, not bigger context windows.</description><pubDate>Thu, 19 Mar 2026 00:00:00 GMT</pubDate><dc:language>en</dc:language><category>agents</category><category>engineering</category><category>ai-native</category><category>rag</category></item><item><title>RAG 101: what it is and why it matters</title><link>https://interlusion.dev/en/blog/rag-101-what-it-is-and-why-it-matters/</link><guid isPermaLink="true">https://interlusion.dev/en/blog/rag-101-what-it-is-and-why-it-matters/</guid><description>A plain-language introduction to Retrieval-Augmented Generation: what it does, how it works, and why it is becoming the default way to use AI on your data.</description><pubDate>Thu, 19 Mar 2026 00:00:00 GMT</pubDate><dc:language>en</dc:language><category>ai-native</category><category>engineering</category><category>rag</category></item><item><title>sqlc: the tool that makes SQL-first development practical</title><link>https://interlusion.dev/en/blog/sqlc-the-tool-that-makes-sql-first-development-practical/</link><guid isPermaLink="true">https://interlusion.dev/en/blog/sqlc-the-tool-that-makes-sql-first-development-practical/</guid><description>sqlc generates type-safe Go code from plain SQL. No ORM, no runtime reflection, no surprises: your schema, your queries, and generated code that compiles.</description><pubDate>Wed, 18 Mar 2026 00:00:00 GMT</pubDate><dc:language>en</dc:language><category>engineering</category><category>go</category><category>tools</category></item><item><title>Autoresearch: the night shift that changes everything</title><link>https://interlusion.dev/en/blog/autoresearch-the-night-shift-that-changes-everything/</link><guid isPermaLink="true">https://interlusion.dev/en/blog/autoresearch-the-night-shift-that-changes-everything/</guid><description>Andrej Karpathy&apos;s autoresearch lets AI agents run about a hundred ML experiments overnight on one GPU. It looks like the beginning of autonomous science.</description><pubDate>Tue, 17 Mar 2026 00:00:00 GMT</pubDate><dc:language>en</dc:language><category>research</category><category>agents</category><category>ai-native</category></item><item><title>Flux vs. ArgoCD: which GitOps tool is built for agents</title><link>https://interlusion.dev/en/blog/flux-vs-argocd-which-gitops-tool-is-built-for-agents/</link><guid isPermaLink="true">https://interlusion.dev/en/blog/flux-vs-argocd-which-gitops-tool-is-built-for-agents/</guid><description>Both are CNCF Graduated, and both do GitOps well. Their architectures disagree about who the primary operator is. That matters once agents enter the loop.</description><pubDate>Mon, 16 Mar 2026 00:00:00 GMT</pubDate><dc:language>en</dc:language><category>engineering</category><category>tools</category><category>ai-native</category></item><item><title>The renaissance of Angular</title><link>https://interlusion.dev/en/blog/the-renaissance-of-angular/</link><guid isPermaLink="true">https://interlusion.dev/en/blog/the-renaissance-of-angular/</guid><description>Angular spent years as the framework people loved to leave. Then it quietly shed everything that made it heavy, added everything that makes it modern, and became the frontend framework best suited for the age of AI.</description><pubDate>Mon, 16 Mar 2026 00:00:00 GMT</pubDate><dc:language>en</dc:language><category>engineering</category><category>opinion</category><category>ui</category></item><item><title>Arc42: the architecture doc template worth keeping</title><link>https://interlusion.dev/en/blog/arc42-the-architecture-doc-template-worth-keeping/</link><guid isPermaLink="true">https://interlusion.dev/en/blog/arc42-the-architecture-doc-template-worth-keeping/</guid><description>Most architecture documentation is either absent or abandoned. Arc42 gives it structure that humans can navigate and AI agents can maintain. Here is how we use it and why it stuck.</description><pubDate>Sat, 14 Mar 2026 00:00:00 GMT</pubDate><dc:language>en</dc:language><category>engineering</category><category>tools</category><category>ai-native</category></item></channel></rss>