How WebstaurantStore used Snowflake Cortex AI to automate dbt staging for 700+ sources in 90 days

Cathy Huang, senior business intelligence engineer at WebstaurantStoreshared how a three-person data engineering team automated a staging layer for more than 700 sources in 90 days, without adding headcount.

The trigger was a legacy SQL Server data warehouse that had hit its limits. The system leaned on more than 1,300 tightly coupled, effectively black-box stored procedures, and as the business grew data volume roughly 40 percent a year, the cracks showed everywhere: overloaded batch jobs risked bleeding past their overnight windows into business hours, engineers hand-coded staging logic table-by-table to keep up, and the opacity of the old stored procedures made automated testing nearly impossible.

The team's fix automates ingestion from landing straight through to a bronze layer. Rather than have engineers manually map cryptic, abbreviated column names from legacy OLTP sources, the pipeline feeds that raw system metadata into Snowflake Cortex LLM functions, which propose normalized, readable column names and business definitions. Those suggestions land in an intermediate review table before they touch anything else. Engineers audit and, when needed, override each one there, and the dbt Jinja macros that actually compile the production model SQL and YAML documentation only ever read from that human-approved table, never from the raw AI output directly.

That checkpoint is what let WebstaurantStore move fast without giving up control: the team onboarded more than 700 sources in 90 days, roughly a tenfold increase in ingestion speed by their own measure, using the same three engineers who built the framework rather than adding headcount. Huang credited a dbt resident architect, Jessica Stayton, with helping the team solve a particularly thorny case: a major source with no watermark to track changes by. As Huang put it, automation only pays off if the people relying on it can still trust what it produces.

Watch every keynote and breakout from dbt Summit 2026 on demand. Watch dbt Summit 2026 on demand →