Skip to content

04. Week 8 — Daily Recall

Spaced practice. Answer from memory. If stuck, jump to the explainer section shown in parentheses.

Monday (after ELI5 + chapter 1)

  1. In one sentence, why does basic RAG stall around 70% on hard tasks? (§1.1)
  2. Walk the Q3-vs-Q4 revenue example and say where retrieval fails first. (§1.2)
  3. Why is “wrong chunk” often worse than “no chunk”? (§1.3)
  4. Module 07 librarian vs Module 08 head researcher — what changed? (ELI5)
  5. Name all five placeholders: rewriter, hypothesis, cross-checker, multi-step plan, confidence gate. (ELI5)

Tuesday (after chapter 2)

  1. Rewrite vs expansion — difference in one clean example. (§2.1)
  2. Decomposition — when is one retrieval call structurally insufficient? (§2.3)
  3. Step-back prompting — what does it ask for first, and why? (§2.4)
  4. Take “How did churn change after the pricing update?” and decompose it verbally. (§2.3)
  5. What is the biggest risk of query rewriting? (§2.1)
  6. Draw the query transformation flow from memory. (§2.5)

Wednesday (after chapter 3)

  1. HyDE — why embed a hypothetical answer instead of the raw query? (§3.1)
  2. When does HyDE help most? When can it hurt? (§3.2)
  3. Parent-child retrieval — what is indexed, and what is returned? (§3.3)
  4. Hybrid retrieval — why keep both dense and sparse signals? (§3.4)
  5. What does RRF do, and why is it convenient? (§3.4)
  6. If the corpus is acronym-heavy, which retrieval upgrade would you try first? (§3.4)

Thursday (after chapter 4)

  1. Retriever vs reranker — what architectural difference creates the quality gap? (§4.1)
  2. Why rerank top-K rather than the entire corpus? (§4.1)
  3. Metadata filtering — give three filters that matter in enterprise search. (§4.3)
  4. MMR — what balance does lambda control? (§4.4)
  5. Why can top-5 by pure score be worse than a diverse top-5? (§4.4)
  6. Draw the retrieve → rerank → filter pipeline from memory. (§4.5)

Friday (after chapter 5)

  1. CRAG — what exactly is being corrected? (§5.1)
  2. Self-RAG — what new behavior appears compared with one-shot retrieval? (§5.2)
  3. What does the confidence gate inspect before allowing an answer? (§5.2)
  4. Give one query that should route to decomposition, not plain rewrite. (§5.4)
  5. Iterative retrieval — what signal tells you to search again? (§5.3)
  6. Why is this module already “agent thinking,” just inside retrieval? (§5.4)

Weekend (cumulative)

  1. List 8 rows from the failure-fix table in explainer §6.1.
  2. Explain the full advanced RAG loop in under 60 seconds. (§6.2)
  3. Answer two interview questions from explainer §6.3 without notes.
  4. Say one production lesson from §6.4 that you would enforce on a real team.
  5. Recite the bridge sentence to Module 09 from §6.6.