#engineering

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.

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?

Prompt Engineering war nur das Vorprogramm. Ob Agenten abdriften oder besser werden, entscheidet Context Engineering: welche Information wann ankommt.

Objekte und Tabellen passen nicht zusammen. ORM-Abfragesprachen schaffen neue Komplexität. Und KI-Agenten schreiben besseres SQL als ORM-Code.

Wenn KI-Agenten API-Backends schreiben und pflegen, zählen andere Kriterien: Vorhersagbarkeit und schnelles Feedback schlagen Eleganz. Das spricht für Go.

Darbit, der Hase im Interlusion-Logo, übers Animiertwerden: biegsame Ohren, maskierte Augen, ein Zylinder, der sich dreht, ohne sich zu bewegen.

Why Interlusion builds its marketing site with Astro, and what a zero-JavaScript-by-default architecture means for performance, simplicity, and AI-native development.

A walkthrough of a three-layer design token architecture with Tailwind CSS v4's native @theme directive. No config files, no build tools, just CSS.

HSL promised human-readable colour and lied about lightness. OKLCH delivers: perceptually uniform colour for P3 displays, design tokens, and AI agents.

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.

VectifyAI'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.

Everyone obsesses over models and prompts. The real bottleneck in RAG is retrieval, and fixing it takes engineering, not bigger context windows.

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.

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.

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.

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.

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.