Sep 2021 – May 2023 · 1 yr 8 mos · India · On-site

• Built ARIMA forecasting models for retail demand planning, reducing quarterly forecast error by 24% and improving inventory allocation decisions across regions for supply teams.
• Engineered PySpark pipelines for large transactional datasets, lowering batch processing time by 70% and enabling faster analytics delivery across business intelligence reporting and operations.
• Trained XGBoost churn prediction models with feature engineering, increasing high-risk customer identification precision by 21% for retention campaigns across subscription products and service lines.
• Performed SQL-based data profiling and outlier detection checks, improving upstream data quality reliability by 27% across reporting pipelines for enterprise analytics and governance teams.
• Applied K-Means clustering to segment customers by behavior, enabling targeted marketing strategies and increasing campaign response rates by 16% across digital channels and regions.

Mar 2021 – Aug 2021 · 5 mos · India · On-site

• Built spaCy NLP pipelines for named entity recognition on support tickets, achieving 86% classification accuracy and reducing manual triage workload across service desks nationwide.
• Implemented Kafka streaming ingestion for real-time user activity data, delivering sub-minute analytical availability for product monitoring use cases across mobile applications and web platforms daily.
• Designed Snowflake data warehouse schemas using star modeling, accelerating analytical query response by 32% for business reporting across finance, marketing, and operations teams enterprise.
• Conducted correlation analysis and ANOVA to identify key drivers of customer satisfaction, guiding quarterly product investment priorities across consumer segments and service offerings portfolio wide.