Order the steps of canonical embed then rerank 2-stage retrieval

Zenaique

Attempt it

1Sort candidates by cross-encoder score and return the top-10 to the consumer
2Pass each of the top-100 candidates to a cross-encoder reranker along with the query
3Encode the user query with the embedding model
4(At index time, separately:) embed every document in the corpus and upsert to the vector DB
5ANN search against the embedded corpus to fetch top-100 candidates

Quick reference

TL;DR

Index documents once, then per query: embed query, ANN top-100, cross-encoder rerank, return top-10.

Memory aid

Sign in for the mnemonic that makes this stick. See Pro pricing.

Easy to grasp

Picture a giant bookshop and a clerk helping you find books. Ahead of time, the clerk has tagged every book with a little label that summarizes what it is about. That tagging is the index-time step. It happens once per book, not once per question. When you walk in with a question, the clerk converts your question into the same kind of label, then quickly grabs a hundred books whose labels look similar. That is the fast first pass. Finally, the clerk reads each of those hundred books a little more carefully alongside your question and picks the ten that actually answer it. That careful reading is the rerank.

Key concepts

Embeddings Vector Databases

Detailed answer & concept explanation

~4 min read

Everything you need to truly understand this topic: intuition, mechanics, step by step explanation, code, formulas, and worked example. Click to expand.

Everything you need to truly understand this topic: intuition, mechanics, step by step explanation, code, formulas, and worked example.

Sign in for 5 free unlocks/day or go Pro for unlimited. See Pro pricing.

Quick scan · 2–4 min

Everything important, quickly.

60s: identify the two timelines, list the five steps in the correct order, explain why index-time precedes query-time and why ANN precedes rerank; name the top-K cutoffs at each boundary.

Real-world examples

Real products, models, and research that use this idea.

Pinecone's reference RAG architecture uses exactly this 5-step shape with OpenAI text-embedding-3-large and Cohere rerank-3.5.
Anthropic's contextual retrieval recipe pairs the same five-step pipeline with Voyage embeddings and a cross-encoder reranker for production RAG.
LlamaIndex's `VectorIndexRetriever` + `SentenceTransformerRerank` post-processor implements the canonical 5-step ordering as the default RAG template.
Weaviate's hybrid retrieval + rerank module exposes the index then query then rerank flow with the same step boundaries documented in 2026 production guides.

Interviewer follow-ups

What an interviewer would ask next. Try answering before peeking at the approach.

QWhat happens if you upgrade the embedding model? Walk through the migration.

Sign in or go Pro to see the approach for this follow-up.

QHow does swapping the dense-only first-stage for hybrid (BM25 + dense) change the pipeline shape?

Sign in or go Pro to see the approach for this follow-up.

QWhen would you replace the two-stage architecture with a late-interaction retriever like ColBERT?

Sign in or go Pro to see the approach for this follow-up.

Don't say this

Red flags and common mistakes that signal junior thinking. Click to expand.

Most common mistake

Treating corpus embedding as a per-query step. Document embeddings are computed once at index time; only the query embedding is computed per request.

Sign in or go Pro to see all red flags and common mistakes. See Pro pricing.

60-second cheat sheet

The night-before-the-interview bullets. Scan these on the way to the call.

Sign in for 5 unlocks/day or go Pro for unlimited revision sheets. See Pro pricing.

References

Primary sources. Skim if you want the original framing.