Meta AI Incognito Chat: What WhatsApp Really Promises
Meta AI on WhatsApp now has an incognito mode that claims to store nothing. We look at what's behind it and what it actually means for you.
Read more →Practice
Flowise: Your Own AI Chatbot, Without a Line of Code
Making the empty vector store speak: with Flowise you build a chatbot that knows your documents by drag-and-drop — Qdrant as the store, a model in front. Including docker-compose, costs, and the road online.
Read more →
Which operating system is right for AI?
Windows, Mac or Linux — what do you actually need for AI? Spoiler: usually nothing new. And there's one annoyance I have to get off my chest.
Read more →
Qdrant: The Vector Database That Just Runs
Taking the vector store into your own hands: start Qdrant with one Docker command, embed documents, search them. Which AI you need for it — and why an embedding costs almost nothing.
Read more →
Prompting: When Is a Prompt Too Long?
„You wanted to buy one beer, you got two." When a prompt grows so long it contradicts itself, you've gone too far. When less is more.
Read more →
Vector Store: How Your AI Gets a Memory
How a vector store works, what RAG has to do with it, and how your AI gets a memory for your own documents — no coding required.
Read more →
Testing AI Models: Eight of Them, One Kebab
Eight AI models, one silly question: why is a PommDöner better than a Lahmacun? This is how you find out which model convinces — and which just waffles.
Read more →
An update knocked out our AI autopilot
A blog about AI, written with AI, stopped by an AI-tool update. The story of a small mishap — and what it reveals about AI tools.
Read more →
Custom AI Models in OpenWebUI — No Code Required
In OpenWebUI you can build your own models: your system prompt, your LLM, your knowledge base — a personal AI assistant that doesn't need re-explaining every single time.
Read more →
Prompt Fail: What AI Won't Figure Out on Its Own
Technically correct, practically useless — or dangerous. Two examples show why AI doesn't read between the lines when you leave something unsaid.
Read more →
Test Your AI Knowledge — Four Levels, Honest Results
Four difficulty levels, 80 questions, zero fluff: our new AI quiz tests your knowledge honestly — and tells you exactly where you stand.
Read more →
AI Scheduler Writes Perfectly – And Fails at the Sandbox
The scheduler ran, the article was good, the commit never arrived. An honest account of AI automation, sandboxing, and the difference between content and infrastructure.
Read more →
Claude Code Built This Blog
A blog about AI, built with AI. Sounds like clickbait, but it's just true. Here's how it actually went.
Read more →
AI Writes My Resume
AI can genuinely help with your resume — if you know how to ask. Step-by-step guide, what works well, and where the pitfalls are.
Read more →