Snowflake

Snowflake sits at the center of more and more customer data stacks — the warehouse of warehouses for both analytical reporting and operational pipelines. When the integration lands, any Snowflake table becomes a queryable source for canvases, functions, and chats. Same SQL surface, same RBAC, no separate query tool.

What you'll be able to do

• Pull live data into a chat — ask "what's revenue by region this month" and get a table from your warehouse without leaving the conversation.
• Build cross-source canvases — Snowflake on one side, BigQuery or Fabric on the other, joined and rendered in a single view.
• Write back from automations — a function can insert rows, run a stored procedure, or update a table when conditions fire elsewhere in your stack.
• Run analytical SQL from chat — the cross-table aggregate that Snowsight handles natively and BI tools handle awkwardly, asked in plain English.
• Time-travel queries for the "what did this look like Tuesday" questions auditors love.

How it'll work

Connect via OAuth or a key-pair credential to a Snowflake user scoped to the warehouses and databases you want exposed. myai uses the standard SQL endpoint — same surface Snowsight and dbt use. Your RBAC carries over unchanged.

Tell us you want Snowflake

Use cases we're hearing

• Executive dashboards built from Snowflake in a single chat-driven canvas — refresh, drill in, ask follow-ups in the same conversation.
• Cross-org reporting for portfolio companies whose subsidiaries land data in different schemas of the same Snowflake account.
• Customer-facing analytics powered by Snowflake under the hood — without standing up a separate BI seat per user.

Reference

• Connection model: Snowflake SQL endpoint (key-pair or OAuth)
• Same auth, same RBAC, same query surface as Snowsight — no parallel permission system to maintain
• Scope: ~35 operations spanning schema introspection, query execution, and warehouse management