Jun 2022 – Oct 2023 · 1 yr 4 mos · Dallas, Texas, United States

Oct 2019 – Jun 2022 · 2 yrs 8 mos · San Francisco Bay Area

Jan 2019 – May 2019 · 4 mos · Dallas-Fort Worth Metroplex

• shopheart.org (American Heart Association)
• Segmented Customer Base using K-means Clustering(Python) and RFM to understand Consumer behavior across channels and developed XGBoost model to predict the probability of customer purchase with 96% accuracy for targeted marketing.
• Developed predictive models(GLM, SVM ,Decision Trees) to predict customer churn and factors contribute to churn. Estimated customer lifetime ,value for different segments to find Optimal Spends using Survival Analysis.
• Built and implemented Random forest machine learning model to predict Zero-dollar walkers (ZDW) to reduce costs associated with fundraising by 2.1 million USD, evaluated important factors that help in the decrease of ZDW.

Aug 2017 – May 2019 · 1 yr 9 mos · Dallas-Fort Worth Metroplex

• Masters in Business Analytics
• > Activities and Societies: Member of Data Science and Big data Club
• > PROJECTS:
• Recommendation Engine – Movie Lens (Scala, Spark MLlib, Apache Zeppelin, AWS EMR)
• Built a movie recommendation engine using Collaborative Filtering on Movie Lens data set with 10 million ratings on Amazon EMR Cluster using Spark MLib. Used Alternating Least Squares (ALS) and matrix factorization (MF) to make personal recommendations for the users in the data set.
• Credit Card Fraud Detection- Kaggle Competition (Pandas, Numpy, Seaborn, Matplotlib, Keras, Tensorflow)
• Built a Deep Learning model to detect and Classify Incoming Credit card Transactions as Fraudulent or Genuine with 99% Accuracy. Used different Sampling Techniques such as SMOTE to increase Recall rate as data has highly Imbalanced Class (500 fraud transactions out of 285,000 Transactions).
• Document Classification Web Application- NLP (python, Flask, AWS Cloud, API, NLP, NLTK)
• Designed and developed an end to end Machine Learning application for Document classification based on Text and deployed it to AWS as an API endpoint using Flask. Built a 3-layer Neural Network for Text classification with log loss of 0.34 and F1 score of 0.86
• Data Visualization- World Bank (Excel, SQL lite, Tableau Prep, Tableau Desktop)
• Developed visualizations, Dashboards, Stories and Animations in Tableau, analyzed reasons behind trends, Patterns, correlations, Evaluated linear and Panel regression and documented findings as a video on World Development Indicators dataset with 2 million records.

Jul 2015 – Aug 2017 · 2 yrs 1 mo · Hyderabad Area, India

Jan 2014 – May 2014 · 4 mos · Hyderabad, Telangana, India

• Responsible for Developing a Machine learning model using Anomaly Detection Algorithm for detecting erroneous behavior.
• Successfully Incorporated all the specified functional requirements and rigorously evaluated the model.