# learning

There are hundreds of videos out there on how to remember everything you read, with millions of views between them. Dan Koe says the whole premise is pointed the wrong way: memory is not an input problem, it is a feedback-system problem. This one covers the four parts of helmsman-style learning, why the second brain turns into a graveyard for ideas, and a note-taking and writing workflow you can actually pull from.

Matt Pocock's senior-programmer recipe sounds like a joke: install /teach, make a folder, paste a prompt. Underneath is a sharp point: AI is eating tactical coding, so strategic learning now needs durable folders that remember progress.

When learning a new tool with Codex, the worst move is asking it to give you a lecture. A better pattern is to ask it for an entry point, a rough map, a tiny exercise, a teach-back check, and breadcrumbs for next time.

Addy Osmani warns that default AI coding workflows help people close tasks, but do not automatically make them sharper. The difference is not whether engineers use AI; it is whether they use it to test and grow their own mental models.

Anthropic's research shows engineers using AI assistance scored 17% lower on tests than those who coded manually. The key difference? Whether they asked 'why' — high scorers used AI to check understanding, low scorers just copied and pasted.