Sofia Calderon

I’m a recent UT Austin graduate with a passion for data science and data engineering!

4 minute read

I just graduated from UT Austin with a B.S. in Computational Biology, a certificate in Programming and Computation, and a minor in Data Science. As I step into this next chapter, I’ve been reflecting on what I’ve learned—not just from textbooks or lectures, but from the process of building, analyzing, and adapting.

📖 Some History

I came into UT Austin as a Neuroscience major despite having done four years of computer science in high school. I was on a bit of…a poorly timed rebellious act, not wanting to follow in footsteps of SWE dad. For some reason my “rebellious” act included wanting to go into healthcare. After some work in some clinics, and finally taking a CS class out of my own volition…I was faced with the fact that I actually love to code!

But I HATED designing webpages, I didn’t want to make video games…so what else is there to do? That’s when I took my first ever statistics class my sophomore spring, and I FELL IN LOVE with data and moreso how coding paired with it. I was the student that was staying after class, going to office hours for fun, and soon enough it led me down to pursue a minor in data science. Each class I took after made me love data and it’s possibilities even further. I became an educated tutor on campus of many of the statistics subject and an “R warrior”.

For much of my undergrad I regretted not coming in as a CS major, but in reflection, I realize that all things happen for a reason and all paths led me to be able to focus on data science courses freely throughout my last two years of undergrad (even if that meant changing to be a computational biology major–because if I’m being honest it was the closest thing to data science before UT had a data science major haha).

📊 Projects That Shaped Me

On the same note, I think it’s worth reflecting on some projects I was able to do throughout my coursework at UT!

🪴 Plant Care Assistant Web App (Fall 2023)

This was a solo Python project where I built a personalized plant care app using Streamlit, the Perennial API, and Pandas. Features included:

  • Real-time data lookup for 5,000+ plant species
  • Automated task tracking (watering, sunlight, care tips)
  • UI designed for beginner-friendly plant support

It taught me how to connect APIs, manage dynamic UIs, and deliver tailored recommendations—an experience close to real product design.

📺 Dr. Who Viewership Analysis (Fall 2023)

This collaborative project focused on time-series data and regressions to understand factors affecting Dr. Who’s viewership across seasons. We:

  • Cleaned and modeled data in R
  • Investigated the effects of showrunners, companions, air dates, and review ratings
  • Used AIC to compare competing models and interactions
  • Visualized viewership and critic score trends with custom ggplot2 visuals

It was a fun and challenging project that taught me how to evaluate models in a social/cultural context and communicate statistical uncertainty clearly.

🎲 World Cube Association Project (Spring 2024)

As a fun capstone, I analyzed data from the World Cube Association to understand performance trends in Rubik’s Cube competitions. I:

  • Cleaned and joined multiple tables of WCA competition data
  • Explored historical patterns in best times and participation by event and country
  • Visualized the evolution of speedcubing over time with animated plots
  • Highlighted gender differences and emerging national leaders in the cubing world

It was a playful but technically rich project that let me practice storytelling with clean visuals, and connect niche interests with broader statistical insights.

🏫 Google School Enrollment Dashboard (Spring 2025)

Another major milestone was designing an education data pipeline using Google BigQuery and Looker Studio. I:

  • Processed 40+ CSVs of K–12 enrollment data across states
  • Wrote SQL queries to examine trends by race, gender, grade, and socioeconomic status
  • Built dashboards to show Section 504 participation and free/reduced lunch rates
  • Applied cleaning and type-casting workflows to make messy data usable

It was my first end-to-end data engineering and visualization project on a cloud platform — and one that directly supported educational equity analysis.

🐦 FeederWatch Bird Project (Spring 2025)

This was one of the most curious and creative projects I’ve worked on. Using 2021 FeederWatch data, I built models to explore how habitat type shapes species richness and abundance. I:

  • Encoded 30+ habitat variables
  • Ran PCA to identify environmental gradients
  • Built decision trees and multinomial logistic models to predict yard type and species
  • Tuned an XGBoost model to impute missing bird counts with solid accuracy

This project taught me how to apply machine learning models to ecological data, and how to translate raw observations into meaningful biological insights.

📚 Courses That Mattered

There were classes that especially challenged and inspired me:

  • Elements of Data Science
  • Elements of Databases
  • Introduction to Statistical Machine Learning
  • Regression Analysis
  • Probability & Statistics
  • Software Engineering and Systems Design

These courses helped build my toolkit: from SQL joins to model cross-validation, from writing unit tests to deploying structured codebases.

💡 Lessons Beyond Code

What I’ve learned isn’t just technical. These projects reminded me that data can be playful, purposeful, and deeply human. From building a plant care app to analyzing bird habitats or visualizing educational access, I’ve seen how data connects us to the world—and to each other. I want to keep exploring problems that matter, designing solutions that feel thoughtful, and making insights that are clear, compassionate, and impactful.

Thanks for reading, and stay tuned — there’s a lot more I’m excited to share.

– Sofia

You May Also Enjoy