Chat With YOUR Notes

Session 4 · Embeddings + RAG · TCE · 50 min · pairs · The weekend's main build.

Runs entirely in your browser (Google Colab). Any laptop or lab PC works — nothing to install, no GPU, no payment. See ZERO-SETUP.

You need: the 2–3 documents you gathered last night (PDF/txt). Forgot? Grab any lecture PDF from your class group right now — 2 minutes.

Part A — Ingest (12 min)

Load your document → chunk (paragraph + overlap) → embed → numpy vector store. Read the Cell 2 sanity print — if the PDF extracted garbage, swap documents now, not at minute 40.

✓ Checkpoint 1: chunk count + vector-store shape printed.

Part B — Search sanity (8 min)

Three test queries about your material. Eyeball the top chunks: right topic? complete thoughts? If not, tune chunk target size and re-run Cells 3–4.

✓ Checkpoint 2: three sane searches shown.

Part C — Full RAG (15 min)

Grounded template (ONLY the context · cite chunks · "I don't know" escape hatch) → ask 5 real questions about your material. Use show_chunks=True — watching retrieval is how you debug RAG.

Part D — Break it honestly (10 min)

1. Ask something not in your docs → must say "I don't know," not invent.
2. Ask something split across two sections → catch the half-truth.

✓ Checkpoint 3: one honest failure + your fix (or why it's genuinely hard).

Stretch

RAG eval — 5 Q + expected through your S2 harness · k=1 vs 3 vs 5 · add your second document · print similarity scores in answers.

Capstone note

Save this notebook. It IS your capstone foundation: S5 (after lunch) bolts tools onto it; S6 attacks and hardens it, then you demo. Choose documents you actually care about — the demo is only as interesting as the data.