TL;DRAI

Query rewriting (multi-query + step-back) improves RAG recall with cheap LLM call, solving vocabulary mismatch between user and documents. For tech leaders: cheap alternative to costly embedding/rerankers reduces latency and infrastructure, influencing budget and stack decisions.

Book: RAG Pocket Guide: Retrieval, Chunking, and Reranking Patterns for Production

Also by me: Thinking in Go (2-book series) — Complete Guide to Go Programming + Hexagonal Architecture in Go

My project: Hermes IDE | GitHub — an IDE for developers who ship with Claude Code and other AI coding tools

Me: xgabriel.com | GitHub

A user types "how do I cancel" into your support bot. Your retriever embeds those three words, runs a cosine search, and hands the model five chunks about cancelling a payment, cancelling a meeting invite, and the word "cancel" appearing in a changelog. The chunk about cancelling a subscription, the thing the user actually wanted, ranked seventh. The model answers from what it got. The answer is wrong, and nobody on your team can see why.

dev.to

Query Rewriting Before Retrieval: The Cheap Recall Win Most Skip

Most RAG pipelines embed the user's raw query and skip the cheapest recall win there is: rewriting it before search.

domenica 14 giugno 2026 New tab

TL;DRAI

1,654 words~8 min read

Book: RAG Pocket Guide: Retrieval, Chunking, and Reranking Patterns for Production

Also by me: Thinking in Go (2-book series) — Complete Guide to Go Programming + Hexagonal Architecture in Go

My project: Hermes IDE | GitHub — an IDE for developers who ship with Claude Code and other AI coding tools

Me: xgabriel.com | GitHub

Query Rewriting Before Retrieval: The Cheap Recall Win Most Skip

Query Rewriting Before Retrieval: The Cheap Recall Win Most Skip

Other newsrooms on this story

Related reading

Metadata Filtering Before Vector Search: The Recall Win Nobody Measures

Context Compression Before the LLM: Cutting Tokens Without Cutting Recall

RAG Rerank: the Highest-Leverage Upgrade to Your Retrieval Pipeline

Search with no AI in the answer, and why I chose plain chunks over tree-RAG

A Cognitive Benchmark for Code-RAG Retrieval: Part 2 — Why Model Rankings…

From Manual RAG to Real Retrieval — Embedding-Based RAG with NVIDIA NIM

Other newsrooms on this story

Related reading

Metadata Filtering Before Vector Search: The Recall Win Nobody Measures

Context Compression Before the LLM: Cutting Tokens Without Cutting Recall

RAG Rerank: the Highest-Leverage Upgrade to Your Retrieval Pipeline

Search with no AI in the answer, and why I chose plain chunks over tree-RAG

A Cognitive Benchmark for Code-RAG Retrieval: Part 2 — Why Model Rankings…

From Manual RAG to Real Retrieval — Embedding-Based RAG with NVIDIA NIM