Train of Thought
By Toria · Published May 6, 2025
Data Explorer · Systems Thinker · Writer in Progress
Photo by SenuScape from Pexels
All Aboard
I’ve been working as a Data Engineer for a few months and we’re building data systems from scratch, but there’s a quiet challenge no one prepares you for:
“How do you show the depth of the work you’ve done… when most of it is invisible?”
This was the question I asked myself when I volunteered to present our team’s progress to the wider department.
To find inspiration I started to look back and what we produced looked like code – lines of logic, joined tables and evolving documentation. But underneath it was a map. A journey. A system.
And one day, while reviewing our original Entity Relationship documents, I saw something strangely familiar.
It looked like the London Underground.
Planning the Route
Each table reminded me of a station and each join was a connection. However, on first glance it looks straightforward like the London Underground map, but then from further inspection you notice the route you want to take is not a direct line and you’re constantly switching lines just to reach your destination.
We wanted to simplify that journey, because just like in real transit systems, it’s smoother to travel up and down a clean route than to keep switching platforms.
In data terms, we wanted to reduce duplication, limit complexity, and design a model that allowed for easy movement. We wanted a model that flowed.
Photo by Jonathan Borba from Pexels
Standing on the Platform
One of our biggest challenges was bridging the old and new — legacy platforms and modern tooling. Which is the difference of travelling on a worn-out train versus a new one. The old train may still get you to your destination but it’s most likely to be loud, slow and has no plug sockets. Compared to the new one that would be a smoother ride, has aircon and would give you space to think.
But even if you move the exact same passengers from one train to another, they won’t sit in the same seats. The data doesn’t transfer perfectly – it needs to be transformed. That’s the phase we found ourselves in: not just carrying data forward, but reshaping it so it fits the future we’re building.
Train Inspection
Once we’d mapped out the journey, I realised something else. We weren’t just designing routes — we were building the actual train.
Like, what even makes a train a train?
What are the essential parts?
What bits are just extra comfort?
What’s outdated, and what should we carry forward?
That’s the phase we’re in now — pulling it all apart and rebuilding it in a way that works better, both for us and the people who’ll eventually use it.
Stepping Off the Train
I’m really glad I stepped out of my comfort zone.
I naturally think in metaphors — they help me see the bigger picture — but this time, it wasn’t just for me. It helped bring people along with the work.
And in doing that, it gave me something too: the nudge to finally start writing, to make sense of what I do in a way that others can see, feel, and maybe even connect with.
Because sometimes, the best way to explain a data model... is to turn it into a train ride.
Written by Toria
Data Constellations — For layered thinkers, quiet disruptors, and curious minds.
Keep exploring the patterns others don’t always see.