Elasticsearch

Depth for interview readiness & durable competence — relevance, internals, performance, Query DSL.

Lessons

0001 Query context vs Filter context
The scored/unscored split — a relevance and performance decision.
done
0002 BM25 scoring: TF saturation & length normalization
How _score is actually computed — three intuitions, one formula, two knobs.
done
0003 Analysis pipeline & the term-mismatch trap
Analyzers, index vs. search time, why queries silently miss.
done
0004 Inverted index internals
Term dictionary, postings lists, and why term lookup is fast.
done
0005 Segments & the merge process
Immutable segments, refresh, merges — and how shards are built from them.
done
0006 The bool query anatomy
must / should / filter / must_not together — interleaves 0001.
done
0007 Aggregations fundamentals
Metric vs bucket aggs — asking the shape of the data, not just matches.
done
0008 Shard sizing & the over-sharding trap
How many shards? — the perf failure mode and how to size them.
done
0009 Mixed retrieval review (interleaved)
A cross-lesson quiz to convert fluency into durable storage strength.
done

These lessons have a teacher attached. Bring a real query or a fuzzy concept to your session and we'll dissect it together.