Quickstart¶

This guide will get you up and running with Strake in under 5 minutes. We will configure a project, add a Postgres source, and run a query using the Python client.

1. Setup Project with `uv`¶

The recommended way to use Strake is with uv.

# Create a new project directory
mkdir strake-demo && cd strake-demo

# Initialize with uv
uv init

# Add the Strake Python client
uv add strake

2. Initialize Strake Configuration¶

Strake uses a declarative configuration file (sources.yaml) to define your data mesh.

Initialize a new configuration using the CLI:

# If you don't have the CLI, install it first:
# curl -sSfL https://strakedata.com/install.sh | sh

strake-cli init

By default, this creates a sources.yaml. Let's ensure it has a PostgreSQL source:

# sources.yaml
sources:
  - name: demo_pg
    type: sql
    dialect: postgres
    # Example connection string
    connection: "postgres://postgres:postgres@localhost:5432/postgres"

3. Query Data (Embedded Mode)¶

You don't need a standalone server to get started. The Python client can run the Strake engine directly.

Create a file main.py:

import strake
import pandas as pd

# 1. Connect (Embedded Mode)
# We pass the path to our config directly
conn = strake.StrakeConnection("./sources.yaml")

print("Connected to Strake Embedded Engine")

# 2. Run a query
# Tables are accessible via: strake.<source_name>.<schema>.<table>
query = """
    SELECT * 
    FROM strake.demo_pg.public.users 
    LIMIT 5
"""

print(f"Running: {query}")
table = conn.sql(query)

# 3. Analyze results as a DataFrame
df = table.to_pandas()
print("\nResults:")
print(df)

4. Run the Script¶

Use uv to run your script with all dependencies managed automatically:

uv run main.py

5. Next Steps¶

Now that you have your first query running:

Add an S3 Source to join S3 Parquet files against your Postgres DB.
Explore the Connectors page for a full list of supported sources.
Check out the Python API for more advanced usage.