HeadsUpAI

Perplexity Computer Connects to Snowflake to Automate Enterprise Data Science

Perplexity launched a Snowflake connector for Computer, its autonomous agent (AI that acts independently to complete tasks). The integration allows users to query live data warehouses in plain language to generate SQL and metrics. It builds on the Perplexity Computer for Enterprise framework by providing a direct pipe into structured internal data.
Availability
Pro, Max, and Enterprise users
Security
User OAuth and Role-Based Access Control
Setup time
Up to 90 minutes for data map generation
Permissions
Read-only access supported
Data privacy
Not used for model training

Enterprise data often remains trapped behind technical barriers. By automating a semantic layer (a map of data relationships) to interpret metrics, Perplexity is applying Perplexity's Agentic RAG research (grounding responses in external data) to business intelligence. This follows a trend of Perplexity's specialized finance agents for high-stakes institutional workflows.

You can now use Computer to forecast demand or analyze customer segments directly from Snowflake. Admins maintain control through User OAuth and Role-Based Access Control, ensuring authorized access only. The connector is available today for Pro, Max, and Enterprise subscribers via the settings menu.

View the full update on perplexity.ai
Perplexity
Perplexity
@perplexity_ai
X

Computer now connects to Snowflake. Run end-to-end work against live warehouse data and get answers with SQL, source tables, filters, and metrics. It’s like a personal data science team, on call with accurate answers from live company data. https://t.co/L1uQC6u5zZ

46retweets471likes
View on X

Still wondering? A few quick answers below.

The Snowflake connector is a new integration for Perplexity Computer that allows users to query live data warehouses using natural language. It acts as a personal data science team by translating plain-language questions into SQL, allowing teams to analyze product usage, customer segments, and business trends without needing technical database expertise.

Once connected, Perplexity generates a data map, which is a semantic layer that teaches the AI how a company's specific data is structured. This map identifies relevant tables and key metrics, enabling the agent to translate natural language questions into accurate SQL queries that run directly against the live Snowflake warehouse to provide evidence-based answers.

The Snowflake connector is available starting May 14, 2026, for users on Perplexity Pro, Max, and Enterprise tiers. It is designed for teams across various departments, including product, growth, marketing, and customer support, who need to make data-driven decisions without waiting for manual reports from a centralized data science team.

No, data accessed through the Snowflake connector is never used to train Perplexity's AI models. The integration is built with enterprise-grade security, including encryption and data privacy protections. It respects existing Snowflake permissions through Role-Based Access Control and User OAuth, ensuring that users can only see data they are already authorized to access.

Admins can enable the integration by navigating to the Connectors section within Perplexity settings. After connecting Snowflake, admins click a button to generate a data map, which can take up to 90 minutes depending on the data's complexity. They can then use a data map editor to review and manage business definitions and shared logic.

Share this update