Data Scientist

Data Scientist — Centric Data Solutions

Chicago, IL | March 2024 – Present

• Engineered and deployed an ensemble machine learning model to predict customer churn, reducing user attrition by 24% over 6 months and retaining $1.2M in annual recurring SaaS revenue.
• Developed a high-throughput real-time data ingestion pipeline using Apache Spark and AWS Kinesis, accelerating feature computation by 42% and processing 15M daily telemetry events.
• Designed and executed 18 randomized A/B tests on user recommendation algorithms, yielding a 14% improvement in conversion rate and a $320,000 increase in quarterly e-commerce sales.
• Automated model deployment workflows with MLOps best practices using Docker and MLflow, reducing average model release cycle times from 14 days to 3 hours with zero pipeline downtime.

Associate Data Scientist — Nexis Insights Group

Boston, MA | June 2021 – February 2024

• Constructed multivariate regression and random forest models to analyze pricing elasticity, optimizing product pricing strategies to increase gross margins by 11% across 4 key segments.
• Implemented automated SQL queries and Tableau dashboards for marketing attribution, saving 15 hours of manual analysis weekly and improving ad spend efficiency by 18% over 9 months.
• Cleaned and preprocessed over 4TB of unstructured customer transaction data using Python and Pandas, reducing data noise by 32% and increasing training dataset accuracy by 15%.
• Collaborated with 3 product managers to define tracking telemetry for new app releases, reducing user onboarding drop-off by 16% through targeted post-onboarding analytics insights.

Python, SQL, R, PySpark, Pandas, TensorFlow, Keras, scikit-learn, Apache Spark, AWS, Docker, Kubernetes, Git, MLflow, Tableau, A/B testing, MLOps, statistical modeling, predictive analytics

Real-Time User Recommendation Engine

Role: Lead Data Scientist

Tools: Python, PySpark, Redis, AWS Kinesis

Designed and deployed a real-time recommendation model for 8M active users, increasing click-through rate by 18% and generating $410,000 in additional quarterly revenue.

Predictive Customer Churn Framework

Role: Data Scientist

Tools: Python, scikit-learn, MLflow, Docker

Built a gradient-boosted classifier to identify high-risk customer churn, reducing subscription attrition by 24% over 6 months and saving $1.2M in annual contract value.

• Certified Analytics Professional (CAP) (2024)
• Google Cloud Professional Data Engineer (2022)

• Languages: English (Native), French (Conversational)
• Volunteer Work: Technical mentor at Chicago Women in Data Science (2022–present)
• Availability: 2 weeks notice