How to connect Snowflake to an AI data analyst
Updated 2026-07-01
An AI data analyst is most useful when it runs on your real data. This guide shows how to connect Snowflake so you can ask questions in plain English and get charts, dashboards and reports back — without writing SQL by hand.
The setup takes a few minutes: connect with a read-only user, let the analyst read the schema, and ask your first question. Your credentials are encrypted, and the analysis runs on your own AI provider key.
- 1
Get your Snowflake connection details
You need your account identifier, a username and password (or key-pair), plus the warehouse, database and schema to use. Copy the account identifier from the Snowflake URL.
- 2
Create a read-only user (recommended)
Create a role with USAGE on the warehouse and database and SELECT on the schema, then grant it to a dedicated user. This keeps the analyst read-only.
- 3
Create your account and add your AI key
Sign up, then add your own AI provider key (bring-your-own-key). If you don't have one yet, a free key takes about a minute to create — your key is encrypted and used only for your requests.
- 4
Add the Snowflake connection
Add a data source, choose Snowflake, and enter the account identifier, user, password, warehouse, database and schema. Test and save.
- 5
Let the AI read and annotate your schema
The analyst annotates your Snowflake tables and columns. With dbt, point it at your target schema so it recognises your models.
- 6
Ask your first question in plain English
Ask something like "Monthly recurring revenue trend this year". On complex questions the analyst shows a readable query plan before it runs anything; review it, then get a chart plus the data behind it. Refine by chatting, then pin it to a dashboard or save it as a report.
Why connect Snowflake instead of exporting
A static export goes stale the moment you download it, and a general AI assistant that only sees a pasted file guesses what your columns mean. A connected AI data analyst reads live Snowflake data, shows the query plan before running it, and remembers what your fields mean between sessions — so answers stay current and consistent.
Snowflake: things to know
- Pick a small warehouse for analytics queries to control credit usage.
- The account identifier is the part of your Snowflake URL before snowflakecomputing.com.
Example questions to ask your Snowflake data
- Monthly recurring revenue trend this year
- Churn rate by plan over the last 6 months
- Pipeline by stage and owner
Keeping it safe
- Connect with a read-only user so the analyst can never change your data.
- A read-only SQL guard blocks anything that isn't a read query.
- Connection details and your AI key are encrypted at rest.
- Your business data isn't stored or used to train any model.
Frequently asked questions
Related
Get your end-to-end AI business intelligence now.
Conversational analytics with AI that understands your data — no SQL, no data team required.