Roshini R

About Roshini R

Data Scientist with 3+ years of experience building predictive models and scalable data pipelines in financial and business analytics domains. Strong expertise in Python, SQL, machine learning, and AWS-based deployments. Proven track record of improving model accuracy, automating ETL workflows, and delivering data-driven insights that improve business decisionmaking.

Experience

1. Data Scientist

   Principal Financial Group

   Jan 2025 — Present · US

   Built and deployed a pension transfer risk prediction model using Python and Scikit-learn, improving prediction accuracy by 35% and enabling better financial risk mitigation strategies.• Engineered scalable ETL pipelines integrating 20M+ customer and transaction records using Python, SQL, and AWS Glue, ensuring high-quality datasets for downstream analytics.• Performed feature engineering and hyperparameter tuning using cross-validation techniques to improve model stability and reduce overfitting.• Deployed production ML models on AWS SageMaker with monitoring dashboards, improving model reliability and maintainability.• Developed executive-level dashboards in Tableau and Power BI to visualize KPIs, customer trends, and operational metrics, improving operational efficiency by 20%.• Built NLP-based sentiment analysis models to analyze customer feedback, supporting targeted marketing campaigns and improving engagement by 18%.• Collaborated with product, engineering, and finance teams to translate business problems into ML solutions and present insights to leadership.• Documented model assumptions, data lineage, and pipeline processes to support governance and reproducibility.

Education

• VEMU Institute of Technology, Chittoor

  Bachelor of Technology , Computer science and engineering

• University of Massachusetts Lowell

  Master of Science - MS, Computer Science