Srinivas Aripirala

Data Science Leadership ► Implemented robust experimentation frameworks using open-source Growthbook for data-driven A/B testing to evaluate recommendation models and optimize digital performance metrics ► Led development of advanced recommender systems utilizing hybrid collaborative and content-based filtering (LightFM), resulting in 13-15% revenue increase per customer ► Directed causal inference initiatives for marketing optimization using Uber's CausalML package, enhancing promotional effectiveness by approximately 20% ► Oversaw customer journey analytics projects that improved email campaign conversion rates by 10-20% through data-driven segmentation and timing strategies Analytics Leadership ► Partnered with Strategy office to develop bottom-up Annual Operating Plan (AOP) forecasts for US eCommerce, ensuring data-driven business planning ► Spearheaded KPI framework development, collaboratively defining metrics, establishing historical benchmarks, and setting targets aligned with AOP objectives ► Led end-to-end analytics infrastructure including Looker dashboard development using LookML, while maintaining ownership of the data pipeline ► Managed product clustering initiatives that drove 70-100% improvement in Google Shopping ads conversion rates by identifying high-potential products Technical Leadership ► Guided forecasting model development for Marketing and eCommerce teams, significantly improving forecast accuracy and business planning capabilities ► Built and managed robust data pipelines utilizing Snowflake, Dagster, and modern Python data stack (Pandas, Polars, NumPy) for efficient analytics workflows ► Implemented MLOps best practices across team projects including model versioning, experiment tracking, and hyperparameter optimization using MLflow, Aim, and Ray Tune

► Successfully ran the biggest Flu vaccination program where we optimized multiple channels of outreach (call, PoS, SMS, etc.). Built multiple deep learning models (architecture based on style GAN) to understand patients' preferences (channel, verbiage, cadence, etc). Flu vaccination program resulted in administering 30% more vaccines YoY ► Championed DataScience Solution for Automation (DoSA) a plugin-based framework, which enabled accelerated data science solutioning by sharing plugins across various Data Science teams ► Optimized Patient adherence levels by targeting the right patients through the right channels at the right time using the right verbiage. Built multiple uplift models using causal forests (pyspark, custom-built algorithm based on causal forests) to maximize enrollment. Uplift models improved the performance by ~80 basis points ► Developed a package for the decision tree based uplift model with parallel implementation. The package not only standardized the usage of the model across the organization but also improved the run-time upto 80% ► Built a Call Volume forecasting model using various Machine Learning models such as Ridge regression Linear, Random Forest and XGBoost. Development led to reduction of the Mean Absolute Percentage Error(MAPE) up to 30% ► Developed and built price optimization model using linear programming. Model deployment led to cost savings of up to $20M

► Developed and implemented various Machine Learning models such as Linear/Non Linear regression models to predict Sales. Implementation led to understanding of relationship between various levers (promotions, seasonality, store labor etc.) with Sales ► Developed and implemented statistical clustering algorithms (k-means, hierarchical clustering) to segment 3300 stores. Analysis led to maximizing the customer service level by understanding customer buying patterns, seasonality, local product preferences, pricing & promotions at a store cluster level

Select Engagement Experiences ► Conducted detailed supply chain assessment to identify inefficiencies for $80B US retail player, using various ML algorithms (Linear/ Non Linear Models, Tree based algorithms etc.) to understand the impact of forecast accuracy, Planogram (POG) analyses etc. Identified annual savings up to $150M ► Realized cost savings by building pharmacist labor optimization model using convex optimization algorithms for a large US Pharmacy retail chain. Identified savings up to 5% in labor costs ► Conducted market assessment, defined unique value proposition by market segment and developed a go-to-market strategy for a US technology firm. Identified annual revenue opportunities up to $75M

Planned, designed, and coordinated integration of new applications and modifications of existing applications of moderate to high complexity ► Developed fully-functional logistics module which integrates and streamlines transportation planning, execution, and freight payment.