T
Tejaswini Pallapothu
Tejaswini Pallapothu
About
Detail
Telangana, India
Experience delivering experimentation-driven ML and decision intelligence: A/B testing & causal inference; ranking/search relevance & recommendations; dynamic optimization (pricing/promotion); return/propensity risk modeling; customer-behavior analytics (funnels/cohorts/path analysis); forecasting/time-series; NLP on unstructured text; plus anomaly detection & imputation for data quality.
Design and ship end-to-end models on Databricks (PySpark/Spark SQL/MLlib), Snowflake (SQL), and Python (pandas, scikit-learn, XGBoost, Prophet), translating behavioral and operational data into measurable KPI uplifts. Collaborate cross-functionally to build reliable feature datasets (batch/streaming/CDC; REST/Kafka; JSON/Parquet/Avro), curate marts (star/snowflake with SCD), and publish self-serve insights in Looker/Tableau/Power BI; productionize with MLflow, feature stores, and CI/CD (GitHub Actions/Azure DevOps/Jenkins), containerized via Docker/Kubernetes with rigorous data validation, privacy-by-design, observability, and cost-efficient scaling.
Design and ship end-to-end models on Databricks (PySpark/Spark SQL/MLlib), Snowflake (SQL), and Python (pandas, scikit-learn, XGBoost, Prophet), translating behavioral and operational data into measurable KPI uplifts. Collaborate cross-functionally to build reliable feature datasets (batch/streaming/CDC; REST/Kafka; JSON/Parquet/Avro), curate marts (star/snowflake with SCD), and publish self-serve insights in Looker/Tableau/Power BI; productionize with MLflow, feature stores, and CI/CD (GitHub Actions/Azure DevOps/Jenkins), containerized via Docker/Kubernetes with rigorous data validation, privacy-by-design, observability, and cost-efficient scaling.
