Phase 2: Dev Environments & Dependency Management · ~30 minutes · Python · uv

"Works on My Machine"

It runs perfectly on your machine because your machine isn't clean. It's the accumulated result of everything you've ever installed, and some of that is invisibly load-bearing.

Hiring signal: Assumes their current setup is one particular accumulated state, not the ground truth, and tests against a genuinely clean environment before trusting it

What you will learn

Explain an implicit dependency: something that 'just works' locally because of accumulated local state, not because it's actually declared anywhere
Explain what reproducibility actually means, beyond 'it works when I run it'
Diagnose a script that fails in a clean environment despite working locally
Identify a likely missing assumption from a clean-environment error message

Introduction

Type: Learn Languages: Python Prerequisites: Phase 01 (Git for Real Work) Time: ~30 minutes

Objective

Learning objectives

Explain an implicit dependency: something that "just works" locally because of accumulated local state, not because it's actually declared anywhere
Explain what reproducibility actually means, beyond "it works when I run it"
Diagnose a script that fails in a clean environment despite working locally
Identify a likely missing assumption from a clean-environment error message

What you're building

Given a script that fails in a clean environment (write your own, following this lesson's fetch_data.py pattern — pick any real package you know is likely already on your machine from earlier work):

Confirm it runs successfully normally (python3 script.py or similar)
Confirm it fails in a genuinely isolated environment (uv run --isolated --no-project script.py), and capture the real error
Diagnose exactly which implicit dependency is missing, based on the error message
Write diagnosis.txt stating the missing dependency and how you'd properly declare it so a fresh environment would satisfy it automatically (this course's Lesson 02-02 covers the actual mechanism in depth)

A script fails on a teammate's fresh machine with FileNotFoundError: [Errno 2] No such file or directory: '/Users/ada/configs/settings.json', even though the exact same script runs fine on the original author's machine. What's the most accurate diagnosis?

The path /Users/ada/configs/settings.json is specific to one person's home directory on one specific machine — it can never exist on anyone else's machine by definition, no matter how correctly set up they are. This is a particularly clear example of an implicit dependency: the script silently assumes a specific file exists at a specific absolute location on the filesystem, without that assumption being declared, configurable, or portable anywhere. The fix is a relative path, a config value, or an environment variable — anything that doesn't hardcode one specific person's machine layout.

Unlock the full lesson

You've read the first 2 sections. The rest of this lesson covers The Problem, Check Yourself, Key Terms & Next — plus a hands-on lab, quiz, and project artifact.

Create a free account to unlock Phase 0 and Phase 1 of every course — no credit card.