Using NotebookLM as a reading layer and Claude as a reasoning layer: does anything get lost?
Alert_Fault_547 · reddit · 2026-10-15
A user with 50-70 local files asks how to let Claude work with them without burning tokens on every query. Their current workflow uses NotebookLM as a reading layer — Gemini extracts information from attached sources — with Claude reasoning over the extracted context. It saves a lot of tokens, but they worry NotebookLM's relevance filtering silently drops important content.
They ask whether Obsidian would be better, whether others use Drive, homemade RAG, or embeddings for the same goal, and how people ensure the LLM actually sees complete information rather than a pre-filtered subset.
More from Apps
- 1Password launches tool letting AI agents log in for you without seeing your password — AndrewSchmidtFC · 2026-10-23
- Lamin goes GA: open-source data platform packs a dozen bio-AI tools into one Python package — ycombinator · 2026-10-23
- Ethan Mollick shares a useful Codex prompt for auditing your computer security — emollick · 2026-10-23
- Escalate details plan for binder-based combinatorial proteomics to feed AI biology models — PatrickKidger · 2026-10-23
- Boardy AI chief of staff refuses to fake data, inspires a smarter market-ranking method — chickefitz · 2026-10-23
- LokalBot ships Linux/Windows alpha: open-source local transcription, dictation and autocomplete — stevyhacker · 2026-10-23