On foundational systems
Good foundations come down to a few decisions: what to simplify, what to standardize, and where to leave room.

By Sherwin Rubio
Enterprise Data Engineer, Newsday Media Group
I’m an Enterprise Data Engineer at Newsday Media Group. I started there in 2008 in cash application, not technology. Today I support about 350 data pipelines and workflows on a platform I designed and built. This is where I write about the decisions underneath that work: what to make reusable, where a team needs its own way of doing things, and how to build something the people who run it can support.
From my writing
From my writing
“Automation is not about replacing jobs but empowering individuals to operate analytically rather than mechanically.”
The approach
How the work multiplies
Much of my work is foundational: reusable patterns, standards, and documentation that other people build on. The aim is to reduce duplication, so teams spend less time rebuilding the same things and more time on their own work. When our team got much smaller, those foundations are what let the people who remained keep everything running.
What I build
Foundations other people build on.
- Reusable patterns for orchestration, validation, logging, and testing
- Shared standards for how work is built, documented, and run
- A shared engineering knowledge base
- Monitoring for process execution, failures, and recovery
What teams get
Less reinvention, more of their own work.
- Faster delivery for the Data Science team
- Simpler onboarding and easier maintenance
- Clear visibility into schedules, failures, and recovery
- Knowledge that transfers between people and teams
What changed
For people first, then for the work.
- Better morale and less stress
- Time for proactive work, instead of more tasks
- The mental space for informed decision-making
As reported in my Redwood Software articles
And it loops back. Time freed from repetitive work goes into proactive improvements, better standards, and documentation, which make the next system easier to build, support, and adopt.
A guiding principle
The principle I work by
“Build for the people you have, not the people you wish you had.”
Get in touch
Working on something similar?
I’m glad to compare notes on data platforms, automation that helps the people doing the work, and standards that teams actually use.
Contact me