• Telemetry Data Insights: Delivering actionable insights by wrangling Forza Motorsport telemetry data, transforming raw data into key recommendations, metrics, and dashboards using Synapse (PySpark, Python, SQL) and Kusto. Created Power BI reports and implemented data pipelines to streamline data analysis, providing insights and data products to stakeholders from Partnerships, Game Design, and User Research teams to improve player engagement and optimize the in-game economy• Player Behavior and Economy Analysis: Built a consolidated player behavioral session table that streamlined data access for teams, enabling efficient analysis of player actions, time spent, and in-game economy metrics. This table also served as a key training dataset for the player segmentation model• Chatbot Development: Built a stakeholder-facing chatbot with embeddings for querying game telemetry data insights. Converted a Synapse notebook-based chatbot into a Streamlit web app, enhancing accessibility and interaction for cross-functional teams, including Partnerships

• Machine Learning Model Development and Deployment: Engineered and deployed machine learning models using Prophet, PySpark, scikit-learn on Databricks, including 300+ time series forecasting models for distinct product classes across 1600 stores, and clustering models to clean survey data• Insights and Visualizations: Managed and fulfilled requests from Finance, Marketing, CRM, Loyalty & Experience teams by working directly with stakeholders and translating their business requests into into data problems and actionable insights. Requests included metrics defining and monitoring, data visualization, deep dive analyses for KPI changes, customer segmentation, customer analytics across channels, etc. Solutions delivered using SQL, Python (pandas or PySpark) and Tableau • Large-scale ETL and Data Modeling: Built Customer 360 data products by integrating and automating various data sources and multiple data science models, including CLV and Churn Risk for 34+ million customers; established a centralized data table serving as the source of truth for all customer loyalty program-related analytics; reduced 50% in turnaround time for ad hoc requests• Automation: Proactively identified opportunities for process improvement, developed and implemented Python scripts to automate data refresh, migration, and email scheduling, saving more than 100 hours of manual or ad hoc labor weekly• Experimental Design and Analysis: Collaborated with Test and Measurement team to design and execute A/B testing experiments, testing the effectiveness of marketing campaigns and personalization efforts. Conducted statistical analyses to assess the impact of these initiatives on driving sales and customer retention

• NLP and Text Classification: Developed text classification Machine Learning models in Python with NLTK and scikit-learn packages with regression and ensemble models, transforming raw customer feedback into actionable insights; enabled the business to accurately identify the products customers complain about and the reasons for these complaints, thereby guiding targeted product improvements across 15K+ stores • Insights and Visualizations: Designed, created, and maintained self-service Power BI dashboards integrating multiple complex data sources for managers and VPs. These dashboards provided critical insights into food safety data and ML-driven analytics, enabling the business to identify and address issues across various parts of the supply chain, resulting in a 36% reduction in customer complaints.• Data Extraction: Developed and implemented ETL pipelines for aggregating and cleansing data from external sources such as Uber Eats, Call Center Systems, Surveys, and distribution centers• Experimental Design: Designed and conducted experiments to evaluate the effectiveness of new food safety solutions in stores, using A/B testing to compare the performance of new solutions against current practices. Conducted statistical test the results to determine their impact on food safety standards

Data Analytics and Algorithm Engineering: Developing algorithms to extract features from massive and messy datasets, conducting experiment design, statistical modeling, time-series forecasting, and visualizing findings and results using Python

• Cleaned, parsed and transformed 75GB raw dataset stored in a MySQL server using R, for the purpose of disambiguating alternate spellings of institutional or individual names from millions of authors and inventors who submit references to genetic sequence database (400k rows of patent data, and 7M rows of reference data)• Designed an innovative algorithm using co-author matching logic that was able to effectively run on large datasets and disambiguate a large portion of data with high confidence

• Being referred to assist teaching a graduate-level data visualization class focused on techniques and data storytelling using R, ggplot2, shiny, D3, and Adobe Illustrator• Mentored students, assessed their work with feedback, lead discussion and Q&A sessions, edited course website, and prepared course materials

• Assisted graduate-level data visualization class focused on data storytelling using R, ggplot2, shiny, D3, and Adobe Illustrator.• Mentored students, assessed work, lead discussion and Q&A sessions, edited course website, and prepared course materials.

• Optimized and monitored keywords and created ads for the pay-per-click campaigns to increase key metrics like acquisition and retention while managing the budget for a mobile app company• Analyzed and reported on paid search marketing data and insights from various marketing channels with statistical analysis, visualization and dashboards, using MS Excel• Drove the daily active users to 2.6 million and total registered users to 30 million by the end of my tenure

• Conducted monthly assessments on psychosocial status of senior participants.• Collected and managed participants' data and medical records. Ensured data quality and data integrity.