Course Wiki

Elasticsearch

The production full-text search engine LLM Zoomcamp references for keyword retrieval at scale.

Elasticsearch appears in LLM Zoomcamp as the industry-standard keyword retrieval engine: an inverted index over document fields, query-time boosting, and scoring — the production-grade version of the minsearch approach the course starts with. The course’s search-evaluation lessons compare candidate retrieval methods, with Elasticsearch as the serious option when a FAQ corpus outgrows in-memory search.

The concepts transfer directly: the course’s keyword-search baseline, hybrid search combinations, and reranking all plug into an Elasticsearch-backed pipeline the same way they plug into minsearch.

Taught in

• LLM Zoomcamp

  • Module 1: Agentic RAG
    • RAG
    • Search
    • Data Ingestion
    • Wrap-up of Part 1
  • Module 2: Vector Search
    • Vector Search with PGVector
    • Next Steps
  • Module 4: Evaluation
    • Next Steps
  • Module 5: Monitoring
    • Next Steps
  • Module 6: Best Practices
    • Document Reranking
    • Hybrid Search with LangChain
    • Next Steps
  • Module 7: End-to-End Project
    • Interface and Ingestion Pipeline ## Related concepts

• Keyword Search

• Hybrid Search

• Vector Search

• RAG

• Reranking