Build maintainable RAG systems through chunking, retrieval, reranking, evaluation, and access filtering.
RAG knowledge engineering
Build maintainable RAG systems through chunking, retrieval, reranking, evaluation, and access filtering.
7 chapters3 casesPublished
LEARNING PATH
Tutorial chapters
Actionable chapters arranged in a useful learning sequence.
chapters · PublishedRAG 基础:文档、分块与数据契约
从可追溯文档管线开始,定义内容、版本、权限和删除语义。
Read chapter chapters · PublishedEmbedding 与向量索引:从实验到可复现检索选择 embedding、索引距离和批处理策略,建立可重建的向量索引。
Read chapter chapters · Published混合检索与重排:提升候选相关性组合关键词、向量和过滤条件,把候选相关性与权限边界分开处理。
Read chapter chapters · PublishedRAG 评测与引用:区分召回问题和生成问题用可追溯问题集检查召回覆盖、引用准确和无依据回答。
Read chapter chapters · Published权限过滤与知识新鲜度把租户权限、文档状态和更新时间作为检索的一等字段。
Read chapter chapters · PublishedRAG 生产管线:更新、缓存与回滚设计可观测的索引发布流程,避免半成品索引和缓存污染答案。
Read chapter chapters · PublishedRAG 查询改写与拒答:先保证证据,再优化表达用受限查询改写和证据阈值处理口语问题、无结果和不确定回答。
Read chapterPRACTICE
Practical cases
Reusable notes built around tools, workflows, and real tasks.
