I Code at Night. I Run Utilities Operations in the Morning.

There’s something almost absurd about it.

You spend the day coordinating potholing operations across two man-made islands in the middle of Hampton Roads. North Island is on the Hampton side, while South Island is on the Norfolk side. You plan the work, support the foremen, and ensure the crews are in the right place, exposing the right lines. Every conflict gets documented properly. Your superintendents manage the ground. You manage the thinking behind it.

Then you come home and write Python.

My Role as a Field Engineer

I’m a Field Engineer II on the Hampton Roads Bridge-Tunnel Expansion. Right now, I’m in utilities, covering both islands and handling the planning and technical support side of potholing operations. The job is about making sure the right questions get asked before anyone digs. Where are the existing underground utilities? Where do they conflict with what’s coming? What does the record actually say versus what reality is going to show us?

You coordinate. You plan. You give the technical direction. Then the foremen and workers execute it, and you ensure everything that comes back gets captured correctly. It’s not glamorous work to explain at a dinner party. But if the planning is wrong, everything downstream pays for it. Every trade that follows is working off the foundation you built.

The Data Dilemma

And somewhere in the middle of running these operations, my data brain started quietly screaming. Because here’s the thing about utilities coordination on a $3.9 billion project: it generates information constantly. Location data. Conflict logs. Depth records. Condition notes. Most of it gets captured manually, lives in silos, and never really talks to anything else.

I’ve seen this pattern before. Not just on job sites, but in businesses. In startups I ran. In analytics projects where the data existed, but the system around it was broken.

Same problem. Different ground!!!

So coding became my evening habit. It wasn’t structured or disciplined in the way a course forces you to be. I just pulled on threads that the day’s work left hanging. Some nights, it’s a script. Other nights, I think through how conflict data could be structured to actually be useful over time. Not just recorded and filed, but queryable. Trend-able. Something you could look back at six months from now and actually learn from.

I’m not shipping anything. Nothing I’m building is polished enough to show anyone yet. But I’m diving deeper into data strategy, automation logic, and how systems should be designed when the inputs are messy and the stakes are genuinely real. That last part matters. The stakes being real.

The Reality of Data Work

There’s a version of data work that happens in controlled environments. Clean datasets. Hypothetical business problems. I’ve done that version. I built crop prediction models at 81% accuracy, ran BI pipelines for businesses, and managed 12,000+ sq km of GIS operations. That work teaches you a lot.

But there’s another version where the problem you’re modeling at 9 PM is something you were planning around at 7 AM. You know exactly how messy the input is because you were the one deciding how it gets collected. Where “automation” isn’t abstract; it’s the thing that would have saved you two hours of manual cross-referencing that day.

That version teaches you differently. It makes you practical in a way you can’t manufacture in a classroom.

A Non-Linear Career Path

My career has never moved in a straight line. Civil Engineering. Drone startup founder. Scaled a company 6x. MS in Business Analytics mid-career. Data and AI consulting. And now, I’m back in the field, on one of the largest active infrastructure projects in U.S. history, coordinating utilities operations between two artificial islands. Each phase looked strange from the outside. This one probably does too.

But the coding at night is how I keep all those layers talking to each other. It’s how I stay sharp on the side of my brain that thinks in systems while the other side is focused on what’s three feet below the surface. Field engineer by day. Data practitioner by night. Not as a strategy, just because the problems I see every day deserve better systems than they currently have.

That’s enough reason ☺️

Building Smarter Solutions

If you’re in the field, any field, and you feel the pull toward building something smarter around the work you’re doing, start with the frustration. Identify the thing that costs you time every single week. Focus on the data that exists but goes nowhere.

Write the script. Structure the record. Build the thing. It doesn’t have to be good. It just has to be honest.

Strategy. Intelligence. Execution. Even underground.

In this journey, I’ve learned that the intersection of data and practical application can lead to transformative outcomes. By embracing the messiness of real-world data, we can create systems that not only streamline our processes but also empower us to make informed decisions.

So, let’s not shy away from the complexities. Instead, let’s lean into them. After all, the future of our work depends on it.