Dallas, Texas, United States
• 5 Years of Industrial Experience • Explored, cleansed and manipulated large structured and unstructured data sets representing millions of users and implemented a different mechanism to impute missing values and data leakage • Designed, developed and implemented end to end ETL packages to prepare data • Performed statistical analysis and visualization to understand data, outliers, and relationships among variables • Implemented feature engineering, feature scaling, PCA and feature crossing for data preprocessing • Designed a/b test experiments and performed hypothesis testing to increase ROI • Designed machine learning models to predict customer life time value which increased ROI by 15% • Performed customer segmentation using clustering techniques to improve targeted marketing • Designed and implemented ML solution to detect fraud which solved critical risk problems • Designed ML models to predict sentiment from user reviews using Natural Language Processing • Implemented advanced machine learning algorithms to increase conversion and predict churn in Tensorflow • Implemented ensemble, regularization, hyperparameter optimization and loss reduction to tune models • Utilized confusion matrix, ROC Curve, gain and lift chart, mse, rmse, gini index to evaluate models • Applied Python libraries such as pandas, sklearn, numpy, matplotlib, tensorflow, keras for ML applications • Used Sqoop to pull data from Hadoop to SQL and vice versa. Utilized Pig for data transformations • Developed Spark scripts using Scala shell commands to perform necessary transformations and actions • Implemented machine learning algorithms such as decision trees, regression, neural networks, clustering to perform quantitative analysis using SAS and Python • Built reporting and self-serve dashboards across platforms to identify user behavior and analyze trends, patterns, and shifts in user behavior, both independently and in collaboration with product owner
- Developed mechanisms to ingest, analyze, validate, normalize and clean data and built high performance data pipelines that generate business ready datasets and insights to support existing use case scenarios - Designed, developed and implemented end to end ETL pipelines that include variables, parameters, expressions, transactions, script task to prepare data for business use cases - Created and maintained OLAP cubes and tabular models for serving different business use cases - Built reporting and self-serve dashboards across platforms to identify user behavior and analyze trends, patterns, and shifts in user behavior, both independently and in collaboration with product owner - Developed efficient SQL queries to build operational stored procedures to solve business problems - Defined site objectives by analyzing user requirements; envisioning system features and functionality. - Provided quality root cause analysis and timely resolution of all the issues reported by the users - Fixed the issues which the end users have been facing within a short span of time. - Proactively communicated with the end users to understand the issue as well as provide workaround steps. - Designed and implemented multiple web based applications using the J2EE frameworks like JSP, servlet, spring, hibernate - Developed android based mobile application which enhanced effectiveness in terms of usability and performance - Coordinated among various other teams for the effective closure of incidents. - Performed deployment activities for critical applications and performed the health check of the applications. - Apart from the assigned task, also came up with innovative ideas and implemented them as utilities which would help the entire team to reduce the effort and thereby enhance the productivity. - Zero escalation from client side. - Provided accurate metrics and adhered to process and compliance.