Data Scientist

I arrive to find a Slack message waiting—the marketing team needs churn predictions by Friday. After my first coffee, I spin up a Jupyter notebook and pull three months of customer behavior data, scanning for missing values and outliers while my mind runs through potential feature engineering approaches. By mid-morning, I'm deep in exploratory analysis, plotting relationships between subscription length and usage frequency. Lunch is eaten at my desk while model training runs in the background. The afternoon splits between tweaking hyperparameters and a thirty-minute call where I walk a product manager through why logistic regression outperforms their favorite black-box algorithm for this particular problem. I create a simple dashboard to show predicted churn risk by cohort. Before leaving, I document my methodology—future me will thank present me. The work feels invisible until it lands on someone's desk and changes what they do tomorrow.