Introduction

I’m Avery Carleen Sigue, and I’ve been working at Switch Automation as a Data Scientist for more than five years. My journey into data science started during the pandemic. It was an uncertain time, but it also pushed me to rethink my career and finally pursue the kind of work I had always wanted to do. I took the risk even though I only had a basic knowledge of Python and SQL.

My role is about turning complex data into something clear, useful, and actionable for the people who rely on it. This involves connecting data from different sources, building and maintaining data integrations, checking data quality, and investigating issues when something doesn’t look right.

In the beginning, Google was my best friend. Whenever I encountered something unfamiliar or ran into a difficult problem, I searched for examples and articles, tried different solutions, made mistakes, and learned along the way. I was also fortunate to have supportive colleagues who took the time to guide me and share their knowledge.

Then AI came along, and suddenly, information felt much easier to access. Research that once took hours could now begin with a single question. To be honest, that scared me a little. Articles and headlines about AI replacing data scientists seemed to be everywhere, and I started wondering what it could mean for my role.

But what if AI isn’t here to replace data scientists, but to become an extra pair of hands that helps us work more efficiently and focus on the problems that need our attention most?

As I began using AI in my day-to-day work, I started to see it differently. Instead of replacing what I do, it became a new best friend that helps make my work easier. AI can now write and understand code, so I use it as a tool to improve scripts, understand unfamiliar code, troubleshoot errors, and spend less time on repetitive tasks.

Some tasks used to take a significant amount of time and consume a lot of memory on my computer. One task required me to validate the last time we received a webhook file to check whether a problem with equipment data was caused by a driver or an API issue. Checking this manually, blob folder by folder, or running a Python driver to do it, required considerable time, effort, and computer memory. With Codex, I was able to show it what to check and how I wanted the results presented. It can now do the repetitive work and give me the information I need. This allows me to find better approaches, automate parts of the process, and solve problems more efficiently.

AI seems to have an answer for almost anything you ask, but that doesn’t mean every answer is reliable or correct. There is still an important role for data scientists. AI-generated code is not always ready to deploy. It still needs to be reviewed, tested, and validated. The logic, context, and final decisions must still come from us, who understand the actual problem we are trying to solve.

Personally, I still want to continue improving my programming skills instead of asking AI to do everything for me, especially when I’m working on a new task. I may ask AI to help me become familiar with the task, understand unfamiliar concepts, or figure out where to start. After that, I still prefer to work through the problem on my own. Once I’ve finished or encountered challenges, I can use AI as a second pair of eyes. After all, nothing quite compares to the satisfaction of solving a difficult problem or creating something new by yourself.

For me, AI is not a replacement for the knowledge and experience I have built over the years. It is an extra pair of hands that helps me learn, solve problems, and work more efficiently. As AI continues to evolve, I also want to keep growing with it, using it thoughtfully while still relying on my own skills, curiosity, and judgment as a data scientist.

About the Author

Avery is a Data Scientist at Switch Automation. She handles data from integrations to customer-ready product solutions.