Project Setup: Environment Setup with Docker, Python, and PostgreSQL

by Yin Yin Chan 🤓 June 2026

TL;DR

This is a continuation to the post on Getting Started: Workflow Setup with Github Codespaces, Visual Studio Code, and Docker.

We prepare our dev environment to build an end-to-end data pipeline with a simple python file structure, PostgreSQL and Docker.

Project files and dockerization
Environment Variables
PostgreSQL in Docker

Project Setup

Prerequisite skills

Basic Python Terminal & Shell Command Line Docker CLI uv psql

Project files and dockerization

Create your pipeline directory

mkdir pipeline
cd pipeline
touch pipeline.py

You can leave pipeline.py empty for now, or check that it works.

# in pipeline.py
import sys

say_what = sys.argv[1]
print(f"Repeat after me: {say_what}")

In your terminal, check that it works

uv run python pipeline.py "Let's get this party started!"

Initialize a python project and check versions

uv init --python=3.14
uv run which python
uv run python -V

Let’s setup a Dockerfile to create the Docker image

# Start an image with python on it
FROM python:3.14.4-slim

# Use uv in Docker 
# (docs: https://docs.astral.sh/uv/guides/integration/docker/)
COPY --from=ghcr.io/astral-sh/uv:latest /uv /bin/

# Set working directory within image
WORKDIR /app

# Activate the project virtual environment to install packages 
# (https://docs.astral.sh/uv/guides/integration/docker/#using-the-environment)
ENV PATH="/app/.venv/bin:$PATH"

# Copy dependency files
COPY pyproject.toml .python-version uv.lock ./

# Install dependencies from lock file
RUN uv sync --locked

# Copy application code 
COPY pipeline.py pipeline.py

# Set entry point
ENTRYPOINT ["python", "pipeline.py"]

Environment Variables

We will use a .env in developement

touch .env

Add to .gitignore so we make sure not to commit it from the go.

# in .gitignore
.env
.env.*
!.env.example

Create your own .env using .env.example as a template for all the key values we’ll need for this project

PostgreSQL in Docker

Create the directory for data persistence

mkdir la_meter_parking_postgres_data

Install PostgreSQL CLI as a development dependency

uv add --dev pgcli

Let’s try it out in terminal. In one terminal window, run:

export $(xargs < .env)

docker run -it \
  -e POSTGRES_USER="$POSTGRES_USER" \
  -e POSTGRES_PASSWORD="$POSTGRES_PASSWORD" \
  -e POSTGRES_DB="$POSTGRES_DB" \
  -v $(pwd)/$DATA_PERSISTENCE_DIR:/var/lib/postgresql \
  -p $PORT:$PORT \
  postgres:18

In another terminal window, run:

#use environment variables, or connect with your values in place
export $(xargs < .env)
uv run pgcli -h localhost -p $PORT -u $POSTGRES_USER -d $POSTGRES_DB

Give your own password and then you should be in. Let’s check

\dt

You should see a blank db schema

Also read

» Data Pipeline: LA Meter Parking Occupancy and Citations with Python and PostgreSQL