LLM Wiki Tutorial - So baut KI dir eine Wissensbasis

There's a fundamental problem with the way we use AI for research and knowledge building. Everything ends up in a chat. Unstructured. Unlinkable. And eventually, the context window overflows, and all the work is lost. Andrej Karpathy has presented a solution: the LLM Wiki. The idea: instead of letting knowledge disappear into a chat, AI builds a real knowledge base from linked Markdown files. Every new source is read, the most important concepts are extracted, existing pages are updated, and new connections are created automatically. In this video, I explain the concept, build an LLM Wiki from scratch with Claude Cowork and Obsidian, and show live how the wiki grows with each new source. What you need: Obsidian (free): https://obsidian.md Claude Code/Cowork or another agent A CLAUDE.md schema file Obsidian Web Clipper: https://chromewebstore.google.com/det... 🎓 My free 5-day course – from prompt typist to systematic AI developer ↪︎ https://mitmario.dev/kurse/ai-coding-... 📧 My newsletter so you don't miss a thing ↪︎ https://mitmario.dev/newsletter 🌟 Benefits Become a channel member and receive exclusive benefits: ↪︎    / @programmierenmario   📌 About this channel and me Web development Efficient and simple. Everything about web development, web design, and related topics. My name is Mario, and I've been working as a web developer for many years. Some tutorials on this channel are suitable for absolute beginners with no prior knowledge, while others are for experienced developers. Get started now!    / @programmierenmario   #ProgrammingM #llmwiki #AI