How Sigma runs a 2,883-model semantic layer on dbt and Dagster
Last edited on Oct 27, 2026
Matt Senick, lead data platform engineer at Sigma Computing, spent his dbt Summit 2026 session walking through how Sigma runs its own semantic layer, the same kind of layer it sells to customers, as versioned code inside a dbt project orchestrated by Dagster.
Senick framed the semantic layer's job in the age of LLMs around three things: giving AI applications guardrails, like PII restrictions and SQL formatting rules, baked directly into model-level instructions; giving every AI surface in the company the same definitions for a given metric so they never disagree with each other; and giving agents enough model context, synonyms, and verified queries to move past generic text-to-SQL guessing.
Borrowing a framing from Anthropic's own published research on the topic, Senick argued that once a question maps cleanly to a defined metric, an agent should be able to call a function and get back the one number every other surface in the company already agrees on.
Inside Sigma's own dbt project, semantic objects are defined and version-controlled directly in SQL and Jinja, mapped to specific warehouse tables, with explicit column-level control over what gets exposed to business users.
Sigma splits its semantic models by business function (sales, marketing, customer success, product, and more) and relies on a small set of dbt packages, including dbt-sigma-data-model, dbt-semantic-view, and dbt-cortex-agent, to compile those definitions into Sigma data models, Snowflake semantic views, and Cortex agents respectively, all as ordinary dbt nodes.
Dagster orchestrates the whole thing: 63 jobs continuously build and deploy across 2,883 dbt models and 6 data regions, downloading the dbt manifest from S3, running a state-aware dbt build, syncing the resulting Sigma data models back through Sigma's API, and re-uploading the manifest for the next run. Senick also previewed Open Semantic Interchange, an effort nicknamed Ossie that aims to let teams convert between a Snowflake semantic view, a dbt metric, and a Sigma data model without locking into any single vendor's format.
Watch every keynote and breakout from dbt Summit 2026 on demand. Watch dbt Summit 2026 on demand →
Get started in dbt
Join the analytics engineers building data infrastructure that actually scales.
Install dbt Wizard CLI
Get started with an agent purpose-built for analytics engineering. It knows which tool to call, which context to pull, and checks its own work before surfacing anything to you.
