As an Enterprise architect, an org of ~1,500 people across 60+ software systems, was responsible for providing Directional/Non-Directional tools to development teams.Created the vision for future state technologies and architecture aligned to the company’s strategic vision for Big data/BI/Database and AI/ML.Helped in strategizing and building POC to engage in capturing business opportunities by providing solutions to business problems.Strategizing with the EA team on BI tool assessment for Looker to reduce the Licensing cost and thinking more futuristic . I was also effectively engaged with the various teams who wanted to be on-boarded to Looker. I led various teams to do POCs in this area under our portfolio. Helping the Looker Platform team in building a Central Looker Platform that can be leveraged across the organization to run and create analytics platforms on.Led the on-prem Hadoop to GCP migration effort for Network System applications as a lead architect. I was involved in a lot of interactions, continual engagement with various groups like Application, Security, Governance, and AI & D amongst few others to come up with current & future architecture in GCP. This gave me an opportunity to demonstrate how we can work across organizational boundaries. Worked with Vendors to understand their products (GCP, DB, BI Tools, GCP logging/Monitoring, OpsGenie ), do hands on PoC and increase the adoption to multiple teams.Led Technical Community of Practice for BigData, AI/ML and BI. I was able to forge this new ground and the success was noticed and is going to be expanded to other portfolios. One of the key outputs from his COP work was our first ever Reference architectures in the area of BigData,AI/ML and BI. This is an ever evolving space and a challenging one as new Technologies emerge everyday. Worked on creating a DB/BigData selection matrix for the team to decide which DB tool is suitable for their application (RDBMS, No-SQL or Big Data)

Health Management Systems, Irving — Machine Learning EngineerJune 2018 - June 2019Project: BRMS - Leading the Business Rule Management System to consolidate all edit rules within multiple systems and platforms into one repository on a Cloud Platform. Working with customers to save millions of dollars a year in overpayment recoveries and millions more through the prevention of improper payments – targeting fraud, waste, and abuse; inappropriate claims; and ineligible claimants.Lead a team of data engineers and analysts in the design, development, and implementation of data pipelines and data modelsWorked on data modeling including feature engineering to achieve better model accuracy. Created machine-learning models with Python, scikit-learn and MLlib to predict reasons for rejected claims for Medicare claims with 85% accuracy.Reduced third-party system costs/processing time by $500K+ per year with machine learning models.Worked on model deployment and maintenance including model scoring, monitoring of model accuracy.Responsible for finding new revenues and improving existing system for information management.Health Management Systems, Irving — Data EngineerJuly 2017 - June 2018Responsible to build the infrastructure required for optimal extraction, transformation, and loading of data from a wide variety of data sources using Hadoop ecosystems.Designed and Developed a highly scalable distributed systems, using Hadoop, Apache NiFi, Sqoop and Hive.Created an automation big data flow system using Apache NiFI to process 10’s of millions of claims on an average for different clients at the same time.Wrote shell scripting to connect to beeline, sqoop, spark and ftp in NiFi.Heavily involved in writing complex Hive scripts to perform ETL.Worked on sqoop to import data from DB2 into Hadoop

Project II: Big Data Solutions: Oct 2016 – PresentData EngineerBusiness Issue:● Ability to analyze the data in real time rather than daily batches● As the company is expanding its market by adding new subsidiaries, existing system is notscaling as it is not meant for horizontal scaling.Goal:● Able to collect and process large amount of data● Ability to store results in persistent storage● A reporting mechanism to view the insights obtained from the analysis● Provide real time monitoring system for Business/Executive.Solution:● Store consultants order information from OE to Kafka. Kafka provides publish subscribe basedmessaging system which will allow other existing systems to read data.● Process data from Kafka to generate required statistics using Spark streaming. SparkStreaming is a fast batch processing system which collects streams of data into small batchand runs batch processing on it. It can handle large amount of data and is very fast inprocessing.● Store the results from spark streaming into Cassandra. Cassandra is a column orienteddatabase which allows existing system to access data.● Access data from Cassandra in Tableau to generate real time reports for Analysis.Responsibilities:Architected and designed a Big data solution to solve batch processing and real time analytic problemsInstalled/Configured Apache Spark 2.0, Scala 2.10, Kafka, Cassandra, and Tableau.Involved in Contest Data Import from MS SQL Server using Sqoop to HDSFDesigned a process to receive data from Order Entry into Apache Kafka.Designed Apache Spark streaming application to pull data from Kafka and update consultant’s ordering data using Spark DataFrames.Applied business validations during Spark streaming processing.Created Cassandra tables to load processed data from streaming.Configured Tableau to provide real time reports and Analysis