Hybrid search for RAG
Mastery: connect the pieces
You can explain, measure, and bound hybrid search for the technical support search assistant without borrowing another topic’s speech.
1Learn the idea
Read
Checklist
- Idea — Keyword and vector search catch different misses. Hybrid fusion ranks candidates from both before generation.…
- Mechanism — Run lexical and dense retrieval, fuse scores (e.g., RRF), optional rerank, then pack evidence for the LLM.…
- Controls — lexical weight, vector weight, RRF k, candidate pools, reranker depth, filters
- Tradeoff — Hybrid raises recall and latency/complexity. Heavy lexical weight helps IDs and hurts paraphrase. Rerankers improve order and cost milliseconds.…
- Failures — ID miss; Fusion drowning
- Metrics — recall@5, MRR, p95 latency, ID-slice recall, citation support
- Ship rule — Ship RRF hybrid with code-preserving analyzer after recall@5 ≥ 0.85 on 50 labeled tickets.
Neighboring layers (retrieval, serving, policy, human review) still own what hybrid search cannot on the technical support search assistant. Do not ask this chapter’s dial to replace permissions or source truth.
Read
Make it operational
Mastery is transfer: take hybrid search into a second scenario that is not the technical support search assistant and rebuild the checklist without copying sentences. If you can only recite this chapter’s examples for hybrid search, you have memorized a story, not a model.
Also pin one numeric memory from this hybrid search chapter: On 50 tickets, pure vector recall@5=0.72, pure BM25=0.64, RRF hybrid=0.86—report denominators. That number is not decoration; it is a template for how claims about hybrid search on the technical support search assistant should look in design docs. Scoped specifically to hybrid search / technical support search assistant / mastery-connect.
Read
Common mix-ups
People confuse hybrid search with neighboring buzzwords when debugging the technical support search assistant. Before changing prompts, ask whether the broken stage was evidence gathering, the hybrid search judgment itself, validation, or the product action. Fixing the wrong stage creates folklore (“we tried hybrid search and it failed”) that blocks the next team on the technical support search assistant. Scoped specifically to hybrid search / technical support search assistant / mastery-connect.
Read
Rehearsal (hybrid-search-rag/mastery-connect)
Write a five-line artifact for this page: goal, inputs, check, owner, stop rule. Invent one fluent failure that the check would catch. Keep details specific to hybrid search rag rather than generic AI advice.
Read
Rehearsal (hybrid-search-rag/mastery-connect)
Write a five-line artifact for this page: goal, inputs, check, owner, stop rule. Invent one fluent failure that the check would catch. Keep details specific to hybrid search rag rather than generic AI advice.
Go deeper
Before you start
Why this matters
Teach hybrid search in ninety seconds using the analogy (two librarians—one who knows exact shelf labels like ERR_CONN_204, one who understands paraphrases like ‘cannot connect’), then replace the analogy with the real artifact names from the hybrid search mechanism page for the technical support search assistant.
In the wild
See how this idea shows up as a product and a company — then come back to the lesson. Skills transfer across vendors.
Related lessons
Check your understanding
Page assessment
Answer from memory. Completion is saved from this evidence, not from opening the next page.
All responses are required.