Guides

AI memory & second brain, made clear.

Why does ChatGPT forget? How does Claude get a memory? What is a second brain with AI? Straight answers, no jargon.

AI Memory: How to Give AI a Long-Term Memory

Why ChatGPT, Claude & co. forget, what AI memory does, and how to spot a good one.

Obsidian, Notion, Claude: what a second brain is, how AI changes it, and why the best second brain maintains itself.

How ChatGPT's memory feature works, how to turn memory on and delete it, and where it hits its limits.

CLAUDE.md, MEMORY.md, the memory feature and MCP: how Claude and Claude Code remember, and how to give Claude a real long-term memory.

Millions of tokens fit in the context window. They're used poorly. Why lean context is the biggest performance boost for your agent.

What an MCP memory server is, how Claude Code and Codex get a memory through it, and what to look for when you choose one.

What a reranker is, how reranking works in RAG, how cross-encoders differ from bi-encoders and when you don't need one.

Vector database or knowledge graph? What graph memory means for LLM agents and why connections are worth more than similarity.

What an AI's context window is, why ChatGPT and Claude forget in long chats, and what context engineering has to do with it.

How AI changes knowledge management in teams and companies, and why AI agents need a shared memory.