SOUMINI REDDY
**************@*****.*** LinkedIn 716-***-**** GitHub USA
WORK EXPERIENCE
Senior Data Analyst (Product & Growth Analytics) ? Epsilon Aug 2025 ? Present
? Analyzed funnel and cohort behavior across 8M+ user sessions, identifying 3 key drop-off points and recommending UX
improvements that were implemented within 2 sprints, contributing to a 10% improvement in retention
? Reduced new user churn by 14% within 3 months by developing an XGBoost model that identified at-risk users 7 days
earlier, enabling timely retention strategies
? Improved experimentation efficiency by developing Python-based A/B testing workflows, helping teams scale from 20 to 120+
tests per quarter while reducing analysis time by 65%
? Developed ETL and data transformation workflows using SQL, Databricks, and dbt to streamline analytics reporting and
improve KPI tracking accuracy
? Collaborated with product and marketing teams on a personalized pricing experiment across 1.2M+ users, using segmentation
and conversion analysis to support changes that improved conversion rates by 18% within one quarter
? Delivered weekly executive readouts through KPI dashboards tracking CTR, DAU/MAU, stickiness, conversion rate, churn, LTV,
ROAS, and funnel performance trends helping stakeholders identify performance trends and support Q3 roadmap planning
Business Analyst ? Amazon Sep 2022 ? May 2024
? Executed complex SQL queries within AWS Redshift to optimize data retrieval and reporting, using methods like JOINs,
GROUP BY, and WINDOW FUNCTIONS to analyze over 10 TB of data and generate insights and reports.
? Conducted deep-dive analysis on operational datasets of over 10 million records using Excel, SQL, applying statistical testing
(t-tests, hypothesis testing) to identify significant performance gaps and validate improvement opportunities
? Partnered with cross-functional stakeholders to gather requirements and present data-backed insights, influencing process
improvements and business outcomes
? Conducted variance analysis to compare expected vs. actual performance, identifying key drivers of deviations and supporting
data-driven business decisions
Data Analyst ? Accenture Jul 2021 ? Aug 2022
? Analyzed customer engagement and campaign performance data using SQL, Python, and Excel to identify user behavior trends
and optimization opportunities across marketing and digital initiatives.
? Developed forecasting and predictive models using Python and historical customer data to analyze engagement trends and
improve planning accuracy by 15%.
? Improved CTR by 110bps over 6 months by generating customer engagement insights through behavioral and feature
performance analysis, partnering with product and engineering teams to reduce decision friction.
? Built interactive dashboards using Power BI and Tableau with advanced calculations (DAX, filters, drilldowns) to track
campaign performance, customer engagement, and business KPIs, reducing reporting turnaround time by 35%
SKILLS
Core: Python (Pandas, Scikit-Learn, Statsmodels), R, SQL
Visualization: Looker, Power BI, Tableau
Cloud & Data Engineering: AWS (S3, Glue, Athena, SageMaker), Azure, Databricks, dbt, Snowflake
Experimentation & Statistics: A/B Testing, Hypothesis Testing, Multivariate Testing, Attribution Analysis, Incrementality Testing
Machine Learning: Feature Engineering, Model Evaluation (Precision, Recall, AUC-ROC), Supervised Learning (Logistic Regression,
Random Forest, XGBoost, Gradient Boosting), Unsupervised Learning (K-Means, Hierarchical Clustering)
Product & Marketing Analytics: Cohort Analysis, Funnel Analysis, Retention Analysis, Customer Segmentation, Conversion
Optimization, Lifecycle Analysis, Behavioral Analytics, User Engagement Analysis, KPI Reporting, Campaign Performance Analysis
Tools: Salesforce, Git/GitHub, Jira, Confluence, Amplitude
EDUCATION
Master of Science, Information Systems - University at Buffalo
Course: Data Structures, Data Warehousing, Predictive Analytics & Machine Learning, Web Analytics, Cloud Computing, Data Science
Bachelor of Commerce, Statistics - Osmania University
PROJECTS
Customer Churn Prediction Project
? Built a predictive model using Python (scikit-learn) on a telecom churn dataset to classify high-risk customers. Engineered features
(RFM, behavioral cohorts, engagement trends) and hyperparameter tuning, handled imbalanced data, and achieved 75-80%
accuracy with logistic regression/XBoost, outperforming baseline models
ETL Pipeline for Analytics
? Built an end-to-end ETL pipeline using Python (Pandas), SQL, AWS S3, and Alteryx to ingest and transform multi-source data
(APIs/CSV), applying business logic (joins, aggregations, data cleaning) and loading curated datasets into Snowflake for analytics
? Automated and orchestrated workflows using AWS Glue, Airflow, and dbt, implementing data validation, incremental loads, and
transformation layers; enabled reliable dashboard refreshes in Tableau and reduced processing time by 40%