Connect Airflow to Snowflake (Beginner Guide)

Overview

In this article, we will discuss how to set up a DAG in Airflow 3.0 that connects to Snowflake, creates a table, inserts some data, then reads that data back.

For a more advanced, production-ready guide covering service users, least-privilege roles/grants, key-pair authentication, and loading data via stages and COPY INTO, check out my advanced guide.

1. Set Up Snowflake

First, we need a Snowflake account and a database to work with.

1. Navigate to signup.snowflake.com and sign up for the free trial.
2. The free trial provides $400 in credits over 30 days (as of January 2026). See Snowflake's documentation for current terms and limitations.
3. Create a new database: YOUR_DATABASE_NAME
4. Inside your database, create a schema: YOUR_SCHEMA_NAME

2. Set Up Airflow Environment

If you already have Airflow configured, feel free to skip to Step 3. For local testing, here's a quick guide to get up and running with Airflow on Docker Compose.

Create the Airflow directory structure

Make a new folder for Airflow and navigate into it:

mkdir airflow-demo && cd airflow-demo

Download the Docker Compose file

curl -LfO 'https://airflow.apache.org/docs/apache-airflow/3.1.6/docker-compose.yaml'

Create required directories

mkdir -p ./dags ./logs ./plugins ./config

Set the Airflow user ID

echo -e "AIRFLOW_UID=$(id -u)" > .env

AIRFLOW__CORE__LOAD_EXAMPLES: 'true'

AIRFLOW__CORE__LOAD_EXAMPLES: 'false'

Initialize the database

docker compose up airflow-init

Start Airflow

docker compose up

Once running, access the Airflow web interface at http://127.0.0.1:8080. The default credentials are airflow / airflow.

3. Configure the Snowflake Connection

Now we need to tell Airflow how to connect to Snowflake.

1. In the Airflow UI, navigate to Admin → Connections
2. Click Add a new record (the + button)
3. Fill out the connection form with your Snowflake details:

Key fields to configure:

• Connection Id: snowflake_demo_test (or your preferred name)
• Connection Type: Snowflake
• Account: use the account identifier shown in Snowsight → ‘Connect’ (do not paste the full https URL).
• Login: Your Snowflake username
• Password: Your Snowflake password
• Schema: YOUR_SCHEMA_NAME
• Database: YOUR_DATABASE_NAME
• Warehouse: Your compute warehouse name

4. Write the DAG

Create a new directory in the dags folder for your project and add the DAG file:

mkdir -p ./dags/snowflake-demo && cd ./dags/snowflake-demo
touch dag.py

Open dag.py in your favorite editor and add the following code:

# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements.  See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership.  The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License.  You may obtain a copy of the License at
#
#   http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied.  See the License for the
# specific language governing permissions and limitations
# under the License.
"""
Simplified Snowflake demo DAG for Airflow tutorials.

Source (Apache-2.0):
https://github.com/apache/airflow/blob/providers-snowflake/6.8.0/providers/snowflake/tests/system/snowflake/example_snowflake.py

Modifications: simplified for readability by Rocky Mountain Tech Lab, 2026-01-15
"""

# Imports
import pendulum
import logging

# Airflow Imports
from airflow.sdk import dag, task
from airflow.providers.common.sql.operators.sql import SQLExecuteQueryOperator

# Constants
SNOWFLAKE_CONN_ID = "snowflake_demo"
SNOWFLAKE_SAMPLE_TABLE = "sample_table"

# ----------------------
# SQL Statements
# ----------------------
# Setup
CREATE_TABLE_SQL = (
    f"CREATE OR REPLACE TRANSIENT TABLE {SNOWFLAKE_SAMPLE_TABLE} (name VARCHAR(250), id INT);"
)

# Sample Insert
INSERT_SQL = f"INSERT INTO {SNOWFLAKE_SAMPLE_TABLE} VALUES ('name', %(id)s)"
SQL_LIST =  [INSERT_SQL % {"id": n} for n in range(10)] # Inserts 0 through 9
SQL_STATEMENTS_JOINED = "; ".join(SQL_LIST)

# Sample Read
READ_SQL = f"SELECT * FROM {SNOWFLAKE_SAMPLE_TABLE} ORDER BY id LIMIT 10;"

@dag(
    'snowflake_sample_dag',
    schedule='@once',
    start_date=pendulum.datetime(2026, 1, 14, tz='UTC'),
    catchup=False,
    tags=['snowflake', 'etl'],
)
def push_to_snowflake():
    # Task Definitions
    create_table = SQLExecuteQueryOperator(
        task_id="create_table_on_snowflake",
        sql=CREATE_TABLE_SQL,
        conn_id=SNOWFLAKE_CONN_ID,
    )

    insert_multiple_records = SQLExecuteQueryOperator(
        task_id="insert_multiple_records_into_snowflake",
        sql=SQL_STATEMENTS_JOINED,
        conn_id=SNOWFLAKE_CONN_ID,
        split_statements=True,
    )

    read_rows = SQLExecuteQueryOperator(
        task_id="read_multiple_records_from_snowflake",
        sql=READ_SQL,
        conn_id=SNOWFLAKE_CONN_ID,
        do_xcom_push=True,
    )

    @task()
    def print_rows(rows):
        logging.info("Snowflake rows: %s", rows)


    # Task Invocations
    create_table >> insert_multiple_records >> read_rows >> print_rows(read_rows.output)

# Dag Invocation
push_to_snowflake()

5. Run and Verify

After saving your DAG, Airflow will automatically detect it (this may take a minute). Trigger the DAG manually from the Airflow UI and watch it execute.

Check the Airflow logs

You should see output similar to this in the task logs:

Verify in Snowflake

You can also verify the data directly in the Snowflake console:

Conclusion

That's it! You've successfully connected Airflow to Snowflake and run your first data pipeline. This basic setup is great for learning and experimentation.

The advanced guide covers production-ready practices including:

• Service user configuration with least-privilege roles and grants
• Key-pair authentication instead of password-based auth
• Loading data via Snowflake stages and COPY INTO

If you have any questions or feedback, please reach out via the contact page.