Mastery: connect the pieces

A complete explanation of RAG pipeline steps has four parts. First: Retrieval-augmented generation first finds external evidence and then asks a model to answer from that evidence. It does not teach the model new weights. Quality is a chain: ingestion, chunking, embedding/indexing, query processing, retrieval, reranking, context assembly, generation, citation, and evaluation. Second, the mechanism: Documents are parsed, normalized, split into addressable chunks, enriched with metadata, embedded, and indexed. At query time, the system may rewrite the question, apply access filters, run vector or hybrid search, rerank candidates, fit the best evidence into a context budget, and generate an answer whose claims point back to sources. Third, the operational controls: chunk size and overlap; embedding model; vector and keyword weights; metadata filters; candidate top-k; reranker depth; context token budget; citation format; abstention rule; and index refresh cadence. Fourth, the evidence: Separate retrieval from generation: recall@k and precision@k for known-relevant chunks; mean reciprocal rank for ordering; answer correctness, faithfulness, citation precision, abstention quality, latency, and cost for the final response.

Use the scenario as an oral exam: For “Can I carry over vacation?”, hybrid search retrieves policy chunks. A reranker promotes the region-specific 2026 policy over a popular 2024 FAQ. The answer quotes the two-day limit and cites the exact chunk; without supporting evidence it abstains. Defend one design choice, then argue against it using this tradeoff: Small chunks retrieve precisely but lose surrounding context; large chunks preserve context but dilute matches. Higher top-k improves recall while adding distractors and tokens. Reranking improves ordering at extra latency. Fresh indexing costs compute but stale evidence damages trust. Finally, identify which of these failures your design catches and which remain: bad parsing, orphaned headings, wrong tenant filters, stale indexes, embedding-version mismatch, low retrieval recall, context stuffed with near-duplicates, prompt instructions inside documents, unsupported synthesis, and citations that point to a source that does not support the claim.

Mastery is not recalling every term. It is predicting consequences before running the system, noticing when evidence contradicts the prediction, and revising the design without moving the goalposts. Keep a decision record containing the workload, baseline, configuration, test set version, results, known limitations, owner, and rollback condition.

For “Can I carry over vacation?”, hybrid search retrieves policy chunks. A reranker promotes the region-specific 2026 policy over a popular 2024 FAQ. The answer quotes the two-day limit and cites the exact chunk; without supporting evidence it abstains.

The worked number is if 18 of 20 test questions retrieve at least one relevant chunk in top 5, recall@5 = 18/20 = 0.90. State the unit and denominator whenever you report it. A percentage without a denominator can conceal a tiny sample; a latency without a percentile can conceal slow users; a similarity score without a labeled task can conceal irrelevant neighbors. Compare the observed value with a threshold chosen before seeing the final test result.

Now test the tempting shortcut. Suppose the team optimizes only the most visible metric. The result may look better while the system becomes less trustworthy. The reason is concrete: Small chunks retrieve precisely but lose surrounding context; large chunks preserve context but dilute matches. Higher top-k improves recall while adding distractors and tokens. Reranking improves ordering at extra latency. Fresh indexing costs compute but stale evidence damages trust. This is why the decision record must include both the intended gain and the tolerated regression. If the tolerated regression is unknown, the change is not ready for a consequential workflow.

• Prefer a measured baseline over a persuasive demo.
• Keep versions, inputs, and thresholds reproducible.
• Separate syntactic success from semantic correctness and authorization.
• Escalate or abstain when evidence falls outside the contract.
• Re-evaluate when data, traffic, models, providers, or user goals change.

These rules turn the topic into an engineering decision rather than a slogan. They also make disagreement productive: another person can challenge the assumptions, rerun the evaluation, and reach a documented conclusion.

Give a three-minute teach-back with no slides. Minute one: define the technique and its boundary. Minute two: trace the mechanism using the worked case and calculation. Minute three: defend the chosen controls with evaluation evidence, then name the strongest unresolved failure. Ask the listener to change one assumption and update your recommendation aloud. You have mastered the topic when the recommendation changes for a technical reason—not because the vocabulary changed—and when you can specify the next experiment that would reduce the most consequential uncertainty.