# Mission: Elasticsearch (depth for interview & professional competence)

## Why
Reach the level where I can explain and reason about Elasticsearch *from first
principles* — not just call the API, but say **why** it behaves the way it does.
The immediate driver is job/interview readiness; the lasting payoff is being the
person on a team who can debug relevance, size a cluster, and design mappings
with confidence.

## Success looks like
- Explain, unprompted, how a full-text query becomes a ranked result set:
  analysis → inverted index → BM25 scoring → shard aggregation.
- Correctly choose **query vs filter context** and justify it on relevance *and*
  performance grounds.
- Reason about shards/replicas, segments, and merges well enough to diagnose a
  slow or over-sharded cluster.
- Write and read non-trivial Query DSL: `bool`, `nested`, multi-field, and
  aggregations — and predict their scoring/performance behavior.

## Constraints
- Learner is **experienced** (already uses ES; comfortable with SQL/databases).
  Skip the "what is a document" basics — aim at the *why* and the edges.
- Practice is a **mix**: real hands-on against a local cluster when it matters,
  in-browser quizzes for recall/theory.
- Timeline is **open-ended** — optimize for durable storage strength, not cramming.

## Out of scope (for now)
- Kibana dashboards, Logstash, Beats (the wider ELK/observability stack).
- Vector search / ML / ES|QL as a primary focus (may revisit later).
- Cloud-provider-specific ops (managed Elastic Cloud tuning).