Second Brain

NotebookLM as a Second Brain

The Story

Charlie’s breakthrough with AI wasn’t a better prompt. It was a better context layer.

He uses Google’s NotebookLM as what he calls a “second brain.” He uploads his own materials: case studies, workflows, past wins, personal stories, customer pain points, YouTube and podcast transcripts, and exported LinkedIn posts. Then, when he asks the AI to generate content, it draws on his actual experience rather than generic knowledge.

The key detail: he sets NotebookLM’s system prompt to speak as him. “You are acting as ME… Always speak as ‘I.’ Never say ‘The author suggests.‘” This prevents the AI from slipping into third-person summary mode, which is the default failure mode of most AI writing tools.

He later articulated the broader principle: “Without context, AI feels clever but fragile. With context, it feels stable, useful, and worth trusting with actual work.” He calls this “context engineering” and argues it matters more than prompt engineering or model selection.

Lesson for Creators

Most people optimize prompts. Charlie optimizes context. The difference: a prompt is a one-time instruction; context is the accumulated environment the AI operates within. When you upload your entire content history, your voice, and your audience’s problems into a tool like NotebookLM, every output starts from a deeper, more specific place. The AI stops sounding like a generic assistant and starts sounding like a research partner who knows your work.

Backlinks