Skip to main content
Airlayer is the in-process semantic engine at the core of the World Model. It compiles your .view.yml and .topic.yml schemas into dialect-specific SQL — resolving joins from entity definitions and protecting against fan-out — so a measure means the same thing everywhere and no query is hand-written or hallucinated. It’s built to be embedded, scripted, or called by agents. Everything the World Model does — the semantic model, the metric tree, the entity graph — is compiled by Airlayer.

What it does

  • Deterministic SQL. Given the same request, Airlayer always emits the same SQL — there’s no LLM in the compilation path, so results are reproducible.
  • Automatic joins. Entity declarations tell Airlayer how views relate, so it builds the join path for you and guards against fan-out double-counting.
  • Dialect-aware. The same definitions compile to BigQuery, Snowflake, Postgres, DuckDB, and more.
  • Embeddable. Airlayer runs in-process as a library or CLI, so agents and apps can compile and run semantic queries without a separate service.

airlayer on GitHub

In-process semantic engine that compiles YAML schemas into dialect-specific SQL. Open source.

Next steps

The semantic model

Define the views Airlayer compiles

The World Model

What Airlayer’s output powers