Niroj Shrestha

About Niroj Shrestha

Analytical and results-driven Data Scientist with 8+ years of experience driving impactful data solutions across fintech, telecom, and healthcare sectors. I specialize in building predictive models, automating ETL pipelines, and delivering actionable insights through intuitive dashboards. My expertise spans machine learning, fraud detection, churn analysis, and cloud data warehousing (AWS, Redshift, Snowflake), with a strong focus on compliance (HIPAA, GDPR, CCPA). I thrive in cross-functional teams, aligning data strategy with business objectives to solve real-world problems at scale.Key Achievements-Deployed fraud detection model reducing false positives by 30% and saving $2M annually in financial losses-Automated ETL pipelines using Python and PySpark, reducing data latency by 60%-Built interactive dashboards in Power BI and Tableau, increasing stakeholder visibility into KPIs and improving decision-making speed-Migrated data infrastructure to Snowflake and AWS, improving scalability and reducing storage costs by 40%-Developed churn prediction model that improved customer retention strategies and reduced churn by 18%-Ensured full compliance with HIPAA, GDPR, and CCPA during all phases of data handling and modeling.

Experience

1. Data Scientist Data Analyst

   Mastercard

   May 2021 — Present · TX, US

   Led advanced analytics initiatives to combat fraud and enhance customer engagement in the financial services space. Built predictive models using Python (Scikit-learn, XGBoost) that significantly reduced false positives and optimized transaction approval rates. Engineered scalable ETL pipelines across AWS and Snowflake, performed in-depth segmentation and A/B testing, and automated NLP-driven sentiment analysis on support tickets. Delivered executive dashboards and data stories that influenced risk, loyalty, and compliance strategies.Key Contributions-Reduced false positives by 27% through fraud detection models using Scikit-learn and XGBoost, improving real-time authorization rates-Conducted advanced EDA and customer segmentation using SQL, Pandas, and Tableau to target high-LTV segments-Developed real-time Power BI dashboards to monitor fraud indices, decline rates, and merchant-level KPIs-Built ETL pipelines with PySpark and Airflow, integrating multi-source data from AWS S3, Redshift, and Snowflake-Automated support ticket triage by deploying NLP models (spaCy, Transformers), reducing manual work by 40%-Ran A/B testing for loyalty offers, using t-tests and chi-square tests to assess campaign lift and conversion impact- Managed model lifecycle via MLflow and Docker, with built-in drift detection and automated retraining-Ensured GDPR and CCPA compliance on all analytics pipelines in partnership with data governance teams-Created executive-facing Tableau visualizations and presentations that shaped product strategy and risk controls.

Education

• Tribhuvan University

  Bachelor of Engineering - BE, Electrical, Electronics and Communications Engineering