Worked on Real time Online Fraud Analytics and built various Machine Learning Algorithms like Logistic Regression, Random Forests etc. Got an opportunity to work on cutting edge technologies like Pyspark, Scala, Kafka, Hive, HDFS and Amazon EC2 cluster.Roles and Responsibilities:o Analyzed and Identified the crucial parameters contributing to the Fraud Mechanism.o Prepared Dashboards using Tableau specifying relationships among Fraud and various Transactional activities. o Accomplished a phenomenal feat of reducing the Fraud by 69%.o Prepared and Cleansed the Data to replicate real world Transactional Data.o Developed a Python script to calculate the aggregated features used in the Financial Domain.o Developed Machine Learning Algorithms to classify a Transaction as Fraudulent or Non-Fraudulent in both Python and Pyspark, Scala.o Created PMML file to be used for open scoring (Prediction).o For a highly imbalanced dataset, used various Sampling Techniques like SMOTE to improve the accuracy of the ML Algorithm.o Created Payment Web Pages to provide real world look and feel of a Transaction to a user.o Led the team of 8 members to automate the entire process and build an end to end solution (For the Online Part).o Achieved an accuracy of 93% with Logistic Regression and 97% with Random Forests.

Worked as a Data mining Engineer (Data Warehousing/Business Intelligence) Roles and Responsibilities:o Analyzed the route operational data to optimize driver’s performance, resulting in 28% increment in profito Analyzed the transactional data of the customers to provide the promotional offers on the basis of their usage history leading to 16% increment in sales.o Implemented a Decision Support System, thereby aiding the decision-making process of the Client. o Developed Standard Operation Procedures to be carried out, in order to follow all the compliances related to the Service Level Agreement. o Constructed a Data Dictionary that explained the Business processes from the niche Waste Industry.o Developed new ETL mappings, sessions, workflows and enhanced the existing ETL's as per the client's requirement.o Managed a critical, Tier 0 and SLA based application.o Conducted Impact Analysis for major change requests and as a part of Defect Management. o Developed an Automation tool which helped in saving 2 man hours on daily basis. o Reduced the Defects/Tickets count by more than 60% (167 to 62).