Cecelia Sciuto
Data-Driven Problem Solver | Data Science & Analytics Enthusiast
Data-Driven Problem Solver | Data Science & Analytics Enthusiast
Hi, I’m Cecelia! I’m an experienced data analyst with 5 years in the field, specializing in sales analytics for the past 3 years. I love turning complex datasets into actionable insights, using SQL for data wrangling and tools like Power BI and Domo to create executive-level sales reports and visualizations. I also enjoy leveraging Python to automate workflows and explore data in creative ways.
Originally from New Hampshire, I moved to Texas to attend Texas State University, where I earned my degree in Finance. After graduation, I lived in Austin for 6 years before relocating to NYC. Outside of work, I’m passionate about traveling and exploring new cultures—especially through food. I’m an outdoor enthusiast and dedicated runner, recently qualifying for the Boston Marathon! When I’m not logging miles or planning my next adventure, you can usually find me hiking with my rescue dog or unwinding with a good book.
This project leverages historical running data from Garmin to predict marathon finish times, tailored to specific training conditions. Using Python and libraries like Pandas, Statsmodels, and Pickle, I performed extensive data cleaning, calculated a custom "power" metric (average speed / average heart rate * 1000), and applied time-series forecasting techniques to create a predictive model.
The model trains on past workout data and requires user input of the race date and expected heart rate to estimate race pace and finish time. This feature allowed me to simulate various effort levels (via different heart rate inputs) to predict how different pacing strategies would impact my performance. The model continuously adapted as I logged new runs, providing real-time insights that I used throughout my marathon training.
This Python-powered calculator was designed to help runners adjust their paces based on real-world weather conditions — specifically temperature and dew point.
Perfect for runners training in non-ideal climates (like Austin's summer heat), the tool lets you:
Input a target pace and convert it to a "warmer" weather-adjusted pace.
OR input a pace you’ve already run in hot/humid conditions and convert it to the equivalent pace for "cooler" (ideal) temperatures.
Just enter your pace, temperature, and dew point, select the direction of the conversion — and get an instant, effort-adjusted pace to guide your training or race planning.
This project connects directly to Garmin’s API to pull real-time running data, which I then analyzed using Python. I focused on visualizing performance trends over time — tracking metrics like pace, heart rate, distance, and training load. The result is a set of clear, insightful charts that helped me monitor my progress and adapt my training strategy with real-world data.