March 31, 2026
From judgment lists to trained Learning to Rank (LTR) models
Learn how to transform judgment lists into training data for Learning To Rank (LTR), design effective features, and interpret what your model learned.
March 5, 2026
Does MCP make search obsolete? Not even close
Explore why search engines and indexed search remain the foundation for scalable, accurate, enterprise-grade AI, even in the age of MCP, federated search, and large context windows.
February 20, 2026
Ensuring semantic precision with minimum score
Improve semantic precision by employing minimum score thresholds. The article includes concrete examples for semantic and hybrid search.
February 17, 2026
An open‑source Hebrew analyzer for Elasticsearch lemmatization
An open-source Elasticsearch 9.x analyzer plugin that improves Hebrew search by lemmatizing tokens in the analysis chain for better recall across Hebrew morphology.
January 30, 2026
Query rewriting strategies for LLMs and search engines to improve results
Exploring query rewriting strategies and explaining how to use the LLM's output to boost the original query's results and maximize search relevance and recall.
All about those chunks, ’bout those chunks, and snippets!
Exploring chunking and snippet extraction for LLMs, highlighting enhancements for identifying the most relevant chunks and snippets to send to models such as rerankers and LLMs.
January 8, 2026
Hybrid search and multistage retrieval in ES|QL
Explore the multistage retrieval capabilities of ES|QL, using FORK and FUSE commands to integrate hybrid search with semantic reranking and native LLM completions.
December 11, 2025
Evaluating search query relevance with judgment lists
Explore how to build judgment lists to objectively evaluate search query relevance and improve performance metrics such as recall, for scalable search testing in Elasticsearch.
December 10, 2025
How to improve e-commerce search relevance with personalized cohort-aware ranking
Improve e-commerce search relevance with explainable, cohort-aware ranking in Elasticsearch. Learn how multiplicative boosting delivers stable, predictable personalization at query time.