NotebookLM
First met in aicp-c3-basecamp-01 · Models & model access · Productivity & source apps
NotebookLMFirst met in aicp-c3-basecamp-01 · Models & model access · Productivity & source apps
Google's personal RAG — drop in up to 50 sources and get cited answers, mind maps and audio overviews over your own material.
You want to learn from documents you already have, with citations, and you don't want to build any retrieval infrastructure.
Every answer cites the exact source and passage — you can jump to where it came from
Generates a two-host audio overview and a video walkthrough of your material
Mind maps the whole corpus, so you see structure you didn't know was there
Ingests PDFs, docs, websites and YouTube videos (transcribed) as sources — 50 free
Answers strictly from those sources, with inline citations
Produces mind maps, audio overviews and video summaries
Doesn't answer from the open internet — that's the point, and the limit
Doesn't help you sell a RAG to clients (Dify is the platform for that)
Doesn't fix a wrong source — if the document is wrong, the answer is wrong
The personal-learning seat: the trainer's daily tool for absorbing research papers. Sits alongside a KB rather than replacing it.
Models & model access · Productivity & source apps
Each course has a NotebookLM notebook; session pages and Path B materials get uploaded as part of the delivery workflow.
Already in the SS-NN delivery checklist (phase B). The mind-map and audio-overview outputs are a free revision layer over material the extraction has just structured.
current
dify: Dify for building sellable RAGs, NotebookLM for personal ones
google-ai-studio: AI Studio to build a prompt, NotebookLM to hold a corpus
| Session | Coverage | Moment | What happened |
|---|---|---|---|
| Basecamp 1: Prompting & RAGs | demonstrated | 2:20:03 | Personal RAG: 50 free sources, cited answers, mind maps, audio/video overviews; trainer's daily learning tool |
| Session 13: Claude Code — Skills, Sub Agents & the Token Economy (posted as 'Beyond Vibe Coding — n8n & More') | demonstrated | 2:25:17 | Explainer video + mind maps from sources; confidentiality caveat for company docs |
| Learning Topics: Sessions 1–5 — Foundations to AI Director | explained | Source-grounded Q&A with citations; 'source-grounded does not mean error-free — open the cited passage' | |
| Session 7: The Outreach Engine | mentioned | 2:49:02 | Recommended study loop: transcript + resources in, interactive deep-conversation audio out — 'have a podcast with me' |
| Session 9: Introduction to Cursor & its principles | mentioned | 0:52:01 | Contrast case: can chat about a video but its transcript isn't available to your agent — harness artifacts must be agent-readable |
| Session 22: Automate Admin Tasks — Advanced RAG (Enterprise 'Internal Perplexity' Built Live) | mentioned | The citation-UX benchmark the final demo is compared against | |
| The Five-Level AI Generalist Roadmap (+ Operating System & Monetization) | mentioned | L1 starter stack and L2 managed knowledge workspace |