Bilt's conversational data layer: How they connected data to LLMs with dbt

Discover how Bilt connected dbt’s Semantic Layer to LLMs, turning data into natural language answers for the business.

Bilt Rewards turned their dbt project into a natural language interface. By connecting their semantic layer and underlying data warehouse to an LLM, business users and data analysts can ask real business questions and get trusted and creative insights. This session shows how they modeled their data for AI, how they kept accuracy intact, and increased data driven conversations across the business.

Speakers:

  • Ben Kramer, Senior Director, Data Analytics at Bilt Rewards
  • James Dorado, VP, Data Analytics at Bilt Rewards
Coalesce 2025: Bilt's conversational data layer: How they connected data to LLMs with dbt | dbt Labs