Ashwini Nadupuri •Email: ada9k0@r.postjobfree.com • Phone: 959-***-****

Summary

Accomplished Data Science and Analytics professional with 5 years of experience in Data Science, Machine Learning and Analytics.

Expertise in design & execution of complex analytical solutions by analyzing business problems, generating business insights using data, developing predictive models, algorithms, interpreting results and recommending strategies.

Proficient in Machine Learning / Statistical model development using R & Python. Expertise in Data Visualization, Metrics Reporting & Storytelling using SQL/HQL &Tableau.

Education

MS - Business Analytics and Project Management - University of Connecticut, GPA 4.075/4 Aug 2020

B.Tech in Electronics and Communications Engineering - GITAM University, GPA 9.11/10 May 2012

Professional Experience

Tata Consultancy Services Jun 2012 to Jul 2019

Data Science and Analytics: Mar 2015 to Jul 2019

Achieved savings of $2.3 million by identifying and targeting customers who are more likely to churn by building Random Forest Classifier in Python. Improved the performance and ensured generalizability (not overfitting) of the built Random Forest Classifier by carrying out Hyper Parameter Tuning using Grid Search with five-fold Cross Validation.

Implemented end to end analytical solution -- including defining analytical business problem, defining population, defining dependent variable, gathering independent variables from various data sources using Hive (HQL), feature engineering, model building & deployment, ROI Analysis and model performance tracking (by building Tableau dashboard).

Carried out Exploratory Data Analysis to understand the associations among variables. Performed data cleaning and feature engineering. Carried out regularization (Lasso Logistic Regression) for feature selection using Python.

Built a sensitivity analysis tool in Tableau and performed sensitivity analysis to decide operational cut offs on the model scores.

Performed A/B testing to establish causal relationship between marketing programs and reduction in churn rates.

Increased the Average Revenue per user of a telecom provider, by 1%, by building Cross Sell Model (Logistic Regression) using R programming to identify and target the customers who are more likely to upgrade their monthly plan.

Created interaction terms based on the insights generated from Decision Trees and included the created features to improve performance of the Logistic Regression Model. Carried out Stepwise Logistic Regression technique for feature selection.

Segmented customers based on their usage characteristics to recommend personalized offer plans using K-Means Clustering.

BI and Reporting: Aug 2014 to Mar 2015

Built customer management Data Mart that contains information on customer call data, plan data, caller network using SQL. Performed several adhoc analysis on user call data to generate business insights and built dashboards using Tableau.

Communicated business insights and results from the analysis to peers, managers and stakeholders by preparing presentations.

Data Base Administrator (DBA): Jun 2012 to Aug 2014

Database planning, installation, creation, upgradation (10g -> 11gR2), patch installation and end to end maintenance.

User and privilege management, tablespace management, data import/export to adhere data retention policy.

Query optimizations to improve efficiency by 20% and provided 24*7 support to business-critical transactions.

Performed data migrations from 9 to 3 databases as a part of consolidation to decommission 9 physical servers and thus contributed in removing inventory wastes.

Academic Projects

Property Insurance Churn Prediction – Predictive Modelling Project:

Built a classification model using Random Forest to predict the customers who are most likely to churn. Performed data cleaning, feature engineering, exploratory data analysis, hyper parameter tuning, feature selection.

Skills

Statistical / Machine Learning: Linear Regression, Logistic Regression, Regularization (Lasso, Ridge), Decision Trees, Random Forest, Gradient Boosting Machine (GBM, XGBoost), Support Vector Machines (SVM), Naive Bayes, K-Nearest Neighbors, K-means Clustering, K – Modes, K – Prototypes, Principal Component Analysis, Time Series Forecasting, Text Mining – Sentiment Analysis, A/B Testing.

Programming and Visualization: SQL, Hadoop (Hive - HQL/Impala), R, Python, Tableau.

Awards and Recognitions

TCS- On the Spot Award 2019

For Building Digital Engagement model that has been effective in reducing marketing campaigning costs.

TCS · Technical Excellence Award 2018

For Building Customer Churn Predictive model that has been effectively identifying the customers who are likely to churn.

Contact this candidate