Sinclaire Schuetze

About Sinclaire Schuetze

I am a Boston-based Data Scientist leveraging advanced predictive/prescriptive models to drive high-value business outcomes in finance and operations. My expertise lies at the intersection of deep learning and economics with a proven track record of optimizing strategy and improving forecasting accuracy in production environments.Current Role: Verizon (Data Scientist Verizon, I operate within the Financial Planning team, focusing on developing and deploying ML models. Key achievements include- Prescriptive Survival Modeling: Built and deployed a production survival model (using GitLab CI/CD and Docker) to optimize loyalty offer targeting, directly contributing to $4M in cost savings by improving retention and repayment- GNN-Enhanced Forecasting: Implemented a Graph Neural Network (GNN) that improved gross adds time-series forecasting accuracy by 16%, strengthening financial KPI predictions- Data Visualization & Strategy: Created a Streamlit dashboard leveraging Teradata and GCP data, providing Finance, Collections, and Marketing teams with critical visibility into offer acceptance and survival KPIs.Research & Advanced ApplicationsMy academic and research work demonstrates a consistent focus on applying cutting-edge techniques to complex, real-world networks- Stanford RegLab: As a Research Fellow, I developed GNN and XGBoost models to classify partnerships by tax noncompliance risk, improving audit targeting across a network of over 7 million entities- Oxford Thesis: My graduate research involved building a Siamese-GNN change-point detection model in PyTorch, which achieved a 0.97 F1 score in identifying economic shocks within global trade networks- Industry Modeling: I significantly enhanced an underwriting model at Mercury Insurance using XGBoost and feature engineering, which increased predicted profit by 28% and accuracy on high-risk policies by 21%.My technical skills include- ML/Deep Learning: PyTorch, TensorFlow, Scikit-Learn, Hugging Face, Graph ML (GNNs), Prescriptive Analytics, Time Series- Platforms & Tools: GCP, Spark, Docker, Git/GitLab, Streamlit- Statistical Techniques: A/B Testing, Causal Inference, Statistical Modeling.I am passionate about applying machine learning to solve financial and operational optimization challenges. Let\'s connect!

Experience

1. Data Scientist Iii

   Verizon

   Jun 2025 — Present · Boston, MA, US

   Built and deployed a prescriptive survival model (GitLab CI/CD, Docker) to score and choose collections customers on loyalty offer acceptance, retention, and repayment.Developed a Streamlit dashboard using data from Teradata and GCP, visualizing offer acceptance and survival KPIs, enabling Finance, Collections, and Marketing teams to monitor performance and drive $4M in cost savings through optimized targeting.Implemented a Graph Neural Network (GNN) that improved gross adds time-series forecasting accuracy by 16%, strengthening financial KPI predictions for the Finance team.

Education

• Wellesley College

  Bachelor of Arts - BA, Data Science and Economics

  2019 — 2023

• University of Oxford

  Master of Science - MS, Social Data Science

  2023 — 2024