Updates both search engineering index files (the headless partial and
the tutorials-lp overview) to list Retrieval Quality Fundamentals and
Building a Golden Query Set alongside the existing Retrieval Quality
Evaluation row. The Evaluation row is also retitled from "Measure
quality and tune HNSW parameters" to "Measure ANN recall and tune
HNSW parameters" to match the refactored page.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Renames precision to recall throughout the ANN-evaluation tutorial so
the page aligns with the ANN-benchmarks convention and with the new
Retrieval Quality Fundamentals page. The numerical formula is
unchanged: when ANN and exact search both return exactly k items,
recall@k and precision@k are numerically identical.
Other changes:
- Remove the Quality metrics subsection, now covered by the
Fundamentals page, and replace it with a short link across.
- Bump weight from 4 to 6 so the three retrieval-quality pages
order as Fundamentals, Golden Query Set, Evaluation.
- Fix a pre-existing prose/code mismatch: the prose said "first
50000 items" while the code uses range(60000).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Introduces two new conceptual tutorials under tutorials-search-engineering:
- Retrieval Quality Fundamentals covers the three-level evaluation
framework (ANN recall, retrieval relevance, business impact), the
evaluation ladder that connects them in practice, and a which-metric-
when decision table keyed by scenario and available ground truth.
- Building a Golden Query Set covers query generation at scale (logs,
LLM synthesis, human annotation) and the failure modes commonly
lumped together as data leakage: synthetic-query unrealism,
embedding-model contamination, near-duplicate documents, temporal
drift, and reviewer reproducibility.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Dylan Couzon