Marimo POC

Marimo notebooks connected to BigQuery, running locally via Docker.

Prerequisites

• Docker Desktop installed and running
• gcloud CLI installed

Usage

./run.sh

The script will:

1. Prompt for your GCP project ID if not set
2. Authenticate to GCP if needed (gcloud auth application-default login)
3. Build the Docker image on first run
4. Start Marimo and open it in your browser

Your notebooks are saved to the ./notebooks/ folder and persist between sessions.

To stop: docker compose down

Querying BigQuery

import marimo as mo
from google.cloud import bigquery
import os

client = bigquery.Client(project=os.environ["GOOGLE_CLOUD_PROJECT"])
df = client.query("SELECT * FROM `project.dataset.table` LIMIT 100").to_dataframe()
mo.ui.table(df)

A working example is available at notebooks/bigquery_example.py.

Updating

git pull
docker compose build
docker compose up