Can You Write Python Without Knowing It? An AI‑Powered Coding Demo

Background

The author, posing as a Python novice, sets out to create a command‑line utility that converts input text into a QR‑code‑like representation displayed directly in the terminal using characters rather than an image file.

Demo

GitHub Copilot is used as the AI coding assistant throughout the demonstration.

Familiarize the Interface

The right‑hand sidebar is opened, the “Project Edit” pane is selected, and the requirement is entered into the input box.

First Conversation

After submitting the request, Copilot generates code suggestions highlighted with a green background. The author accepts the suggestion and runs the program, encountering an error because a required dependency is missing.

Feeding the Error Back

The error message is copied into the “Project Edit” pane, prompting Copilot to modify the code. Copilot adds an import for the missing package and suggests installing it.

Installing the Dependency

Even without prior knowledge of Python package installation, the author invokes Copilot’s inline chat to generate the appropriate pip install command, runs it, and confirms the package installs successfully.

Iterating on Output

Running the program again yields output that is longer than the terminal width. The author informs Copilot of this issue; Copilot adjusts pixel‑size parameters (changing a value from 5 to 2, then to 1) to reduce output width. After accepting the changes, the program produces a rough QR‑code‑like image using hash characters.

Refining Characters

To make the visual closer to a traditional QR code, the author asks Copilot to replace the hash characters with block characters. Copilot updates the code accordingly, and the final run displays the desired block‑based QR representation.

Summary

The case study demonstrates that even users with no Python experience can leverage an AI coding tool to rapidly develop a functional CLI application, while also highlighting typical obstacles—such as missing dependencies and output formatting—and how iterative prompting resolves them.