overview
- Developed Marketing mix modeling to optimize marketing spend across different channels, boosting ROI by 6
- Leveraged marketing analytics and hypothesis testing using Python (Pandas and NumPy) to provide business
- Performed feature engineering, linear regression analysis, and optimized re-allocation using statistical
- Employed multi-touch attribution with deep learning techniques to assess the impact of different customer
- Performed data collection and processing of structured and unstructured data from multiple data sources, like
- Facebook Ad Manager, Google AdSense, APIs, and relational & non-relational databases for data integration
- Built causal inference based uplift modeling using propensity score matching and SK Learn to deliver actionable
- Utilized Tree-based algorithms like Random Forest, XGBoost, and neural nets using Scikit learn and PyTorch for lead scoring and increased patron customers by 7% using A/B testing
- Conducted exploratory data analysis to identify trends and patterns, used dimensionality reduction & feature
- Implemented K-means clustering, and statistical tests for clustering using Scikit Learn, Keras, and TensorFlow
- Calculated customer lifetime value (CLV) with customer churn using time series forecasting models like ARIMA
- SARIMA, and FbProphet on PySpark and AWS EMR
- Leveraged statistical tests for model validation and model optimization through hyperparameter tuning
- Collaborated cross-functionally to create interactive dashboards for data visualizations and tracking Key
- Performance Indicators (KPIs) using Tableau
- Environment: AWS S3 bucket, EMR, Redshift, Athena, Glue, Lambda, Catalog, Sagemaker, PySpark, Tableau, SQL