My first time using Python at work

Yesterday, I used Python at work for the first time.

Before starting my current data analyst role, I spent a few months learning how to program—in Python, specifically. I built some cool things, like a Spotify Popularity algorithm (that only worked for my playlists) and a way to automate my learning (that worked locally but I didn’t have the skills to put it in the cloud). Neither of these, nor anything else I built, had real-world value, sadly.

Still, I enjoyed Python. At my peak Python journey, I coded for 5+ hours a day. That’s because I only worked part time. Once I started my data analyst job, I had less free time. Instead of leverging that time coding, I spent most of it writing on my blog and LinkedIn. Python faded out of my life. I stopped coding.

Looking at my GitHub page now, I see that I had 169 contributions between May 29th and October 6th of 2023. Then, I had zero contributions in 2024. The next one came recently, on January 2nd, 2025.

I didn’t write a single line of Python or any other language for over one year. I only began coding again a few weeks ago on December 22nd, 2024. But I didn’t use GitHub for the first two weeks, hence why 2024 had zero contributions.

Part of me wishes I would have continued to code. I’m sure I would have created some incredible applications and tools during 2024.

But I can’t change the past. I can only make up for it in the future.

Now, I’m revisiting Python. I have so far built a tool to automate my currently non-existent file backup process. In addition to this, I’m diving into web development. One of my goals for 2025 is to rebuild this website from scratch. So, I’m currently learning HTML, CSS, and JavaScript to make that happen.

Anyway, back to Python…

If you have spent any time in the data space on LinkedIn, you’d know that many data analysts and engineers are obsesed with Python. “Pandas” is the one word that makes their knees weak and minds vulnerable. Many data professionals cannot get enough of analyzing data in Python over SQL and Excel. After spending the last year using SQL every day at work, I doubt Python is better. That’s a story for another time, however.

Over the 15+ months I have been a data analyst, I have never used Python at work. Until yesterday…

But I didn’t use the pandas library to analyze a dataset. Nor did I use requests to pull data from an api. In fact, I did not use it for data at all. Instead, I used Python to create a pretty math equation. I know, doesn’t sound so exciting. That doesn’t mean it’s not valuable, however.

Here’s what I created…

It isn’t anything fancy. It might even confuse you. But for my team and I aw work, this makes perfect sense. Let me give you some background info to make this make sense.

Part of my role as a data analyst involves supporting our software development team with data modeling and database management. This often involves creating/managing SQL views and stored procedures.

For the project in question, my team was tasked with altering a view that returns data regarding information costs for our products. (We manufacture, package, and sell vitamins, in case you’re wondering)

Previously, the screen displaying our data only showed the information in terms of stock unit of measures. But it turned out many users would prefer to view this data in terms of production unit of measures. So, my team had to change the view’s underlying schema to provide our software engineers with the requested data.

And I was the lucky one given a chance to make this simple change.

Sadly, no change is ever simple. The conversion, which should have been as simple as multiplying by some conversion factor, turned out to be a nightmare. For reasons I won’t get into here (one because I don’t want to relive the nightmare and two because I don’t want to explain all of it), I couldn’t use any multiply conversion factor.

Instead, I had to bring out my math skills and write a formula for us to determine our item ingredient’s production standard cost. I started to write out ideas in SQL and test the view. Eventually, after trial and error, I came across an equation that worked.

The only problem, however, was that I didn’t know how to explain it with words. All the logic made sense in my head, but I couldn’t just alter the view, deploy the changes, and tell everyone, “Trust me; it works.” That wouldn’t have been wise.

I knew I would have to explain explain to my team my propsed change and why it worked. So, I decided to create a simple visual to explain my logic.

This is when I remembered some old posts from my former blog (RIP Dorky Data). I wrote a lot about math concepts I revisited for the first time since college, and many of those posts included screenshots like the one below:

‍

The above equation has the same formatting and special symbols as you would see in a textbook. You can (somewhat) easily format text like this using LaTex. I forgot which software I used in the past to create images like this. But since I started programming again, I figured there had to be a Python library to visualize math equations.

After using Microsoft CoPilot to conduct research, I discovered the SymPy library could do what I needed.

Since I had other tasks to do at work, I didn’t want to spend too much time reading through documentation or StackOverflow to create what I needed. So I again used CoPilot to generate code I could use as a template. After a few back and forth prompts for clarification, I learned how to write the code I needed. And a few minutes later, I ended up with the above image thanks to the below code…

I realize the image may be a bit blurry. Click it to view the content more clearly. (I would link to my GitHub here, but this code is in a private repository, unfortunately)

The code turns out to be fairly simple. And nothing I did here was anything fancy or complex with Python. All I did was notice a problem (that I needed a visual to explain a solution I had), recognize a potential solution (that Python could help), and conduct research to implement the solution.

The final result didn’t have to be top quality. All I needed was something easy enough for my team to understand, and that’s what I got out of this little Python side quest.

So, I count this as a win.