What This Is

These are lessons I’ve learned solving real problems at QVC. They’re written from my experience — mostly Python, dbt, and DuckDB — but the underlying principles often transfer to other tools.

If you use R, Excel, or something else entirely, the concepts might still help even if the code doesn’t.

The Guides

GuideProblem It Solves
Automating Tableau ExtractsHow to publish data to Tableau Server without opening Tableau Desktop
Building a Transformation LayerHow to structure transformations so you’re not starting from scratch every time
Working with Large Data LocallyHow to handle datasets that push against your laptop’s RAM limits

More guides coming as I document what I’ve learned.


How These Are Structured

Each guide follows a simple format:

  1. The Problem — What you’re trying to solve
  2. Why It Matters — Why the obvious approach doesn’t work
  3. The Solution — What I did
  4. The Code — Examples you can adapt
  5. Related — Links to deeper context in the Stack or Journey

A Note on Tool Choices

I use Python, dbt, and DuckDB because they solved my specific constraints (no cloud access, 16GB laptop, read-only Oracle). Your constraints might be different.

Where possible, I’ll note alternatives:

  • If you’re using R, DuckDB has an R package
  • If you’re using Excel/Power Query, some patterns translate
  • If you have cloud access, you probably have better options than I did

The goal is to share what worked, not to prescribe what you should use.