TAMANNA SHARMA

Developed machine learning-driven propensity models utilizing classification algorithms for lead prioritization, enhancing CTR and reducing COA in personal loan and credit card business verticals. Implemented RFM and bureau data-based segmentation, optimizing marketing campaigns by reducing COA by 1% (month-on-month) and increasing customer engagement. Developed a tailored recommendation engine utilizing advanced ML techniques for website banners, boosting user engagement and conversion rates by offering highly relevant loan and credit card suggestions tailored to individual needs. Employed association rule learning algorithms to generate CleverTap segments for outbound and remarketing campaigns, enhancing targeting efficiency and customer reach. Implemented Python ETL scripts for automated data retrieval, integrating into EDW and multiple destinations. Migrated existing EDW and Python codebases to PySpark, optimizing performance and scalability within Databricks and Data Lake environments for data science and ML applications at big tech product companies.

In the Repeat NCE Metrics project at Lam Research, partnered with business groups to conduct cohort analysis, achieving a 16% reduction in customer complaints within 4 months. Within the eTTTM project at Samsung, collaborated with key stakeholders to establish business processes, streamlining report delivery SLA from 7 days to approximately 2 days.

Leveraged PowerBI and Excel for data-driven refinement of production processes, achieving a reduction in b-grade cells from 1% to 0.7% per day and a decrease in breakage from 0.53% to 0.21% by employing robust statistical analysis techniques.