I thought I was becoming a Data Engineer.

Last year, I spent months grinding through YouTube tutorials, setting up cloud services, and clicking around in Azure Data Factory.

On paper, I was “learning” Data Engineering.

But I wasn’t.

The problem?

I was focusing on tools instead of skills.

I could build pipelines in Azure, but I didn’t know how data actually worked.
I could set up Spark clusters, but I didn’t understand basic SQL.
I was deploying infrastructure, but I wasn’t engineering data.

And when something broke?
I had no clue how to fix it.

That’s when I realized the mistake most people (including me) make:

→ They chase the latest tools instead of mastering the fundamentals.

The truth is, Data Engineering is a software discipline first.

If I had to start over, I’d forget about cloud and focus on three things:
• SQL (because 90% of real-world data lives in databases)
• Python (because it’s the language behind everything from AI to ETL)
• ETL/ELT (because moving and transforming data is the core job)

Only after that would I touch cloud platforms.

I break this all down in my latest video, including:
• Why the "cloud-first" approach wasted months of my time
• The exact roadmap I’d follow if I were starting today
• How to learn Data Engineering the right way (without tutorial hell)

Watch it here → https://youtu.be/N7k70DTC42w

If you’re struggling to piece everything together, I also recommend this structured learning path:

👉 DataCamp’s Data Engineer in Python Track (50% off)

It’s the roadmap I wish I had when I started—hands-on SQL, Python, ETL, and real-world projects instead of just clicking around in cloud dashboards.

Luke