• Design, implement and build ETL pipelines that deliver data with measurable quality• Understand the data needs, including schema, frequency of the data (either streaming, daily, hourly etc) by interacting with machine learning engineers and data scientists• Develop Scala (or equivalent) scripts to generate enriched tables or logs using distributed scalable platforms (Databricks)• Interact with client engineers to discuss granularity of details of the request

• Build and manage financial processing jobs and reports.• Work directly with product stakeholders and contribute to overall big data platform development.• Work on emerging technologies and tools for data analytics. • Design and build big data pipeline that can transfer and process several terabytes of data using Apache Spark, Kafka, Python, Hive and Impala.

• Spark 2.2 deployed on Databricks Entreprise.• KPI metrics for international clients.• Rebuilt ETL pipeline to run in parallel. Improving existing ETL by ~50% to upwards of 90%.• Wrote utility frameworks for Cleaning, Normalization, and Validation in Scala 2.11.• Also wrote multiple utility functions for running parallel notebooks using multiple executors using ThreadPoolExecutors.• Used functional programming and higher order functions for making the code more scalable.

A comprehensive data, intelligence and collaboration platform for firms servicing, targeting and investing inhedge funds.1) Worked on ingestion of SEC form data.2) Address Geocoding exploring Google, Bing, and Here API’s.3) Ingestion of Barclay Hedge Data.Tech Used: Python, Airflow for workflow ETL pipelines, Hive, Hue/Beeline, Pandas data frames, Docker for local testing, Sql Server, Geocoder API, Bamboo for CI, JIRA, Bitbucket.

Devops project using ansible to start-stop and install an apache storm on a 3 node cluster.

* Teaching a lab session every week on a programming course Python.* Interacting with students and solving their doubts during office hours.* Correcting midterm papers, final's papers and weekly assignments.

* Worked on moving computationally heavy operations of the robot like motion planning and image processing onto the cloud.* Worked on Apache Storm and on Python * Used FutureGrid which is a project led by Indiana University and funded by the National Science Foundation (NSF) to develop a high-performance grid test bed that lets scientists collaboratively develop and test innovative approaches to parallel, grid, and cloud computing.* Had to collaborate with 2 other departments i.e. the Computer Vision team and the Cloud Computing team.* Also worked on modularization of the robot's code which was written in Python.

* A general resource reservation framework for cloud computing frameworks while also focusing on bare metal resource provisioning. It is important that resource reservations are supported in a multi-user, multi-project environment. Worked with a team that developed a system in which users will have access to resources based on time slices. The interface to the framework allows access through command line tool, Web interface, as well as a programming API in python. By the end of summer I was able to start and shutdown Virtual machines using this reservation framework successfully. * Was also a mentor to a student who had come to Indiana University for a Summer program from Dillard University New Orleans. I mentored the student on Python and MongoDB for the Cloud computing project.* Technology Used: Flask, Docopts, Python, Openstack-Havana, Rackspace, Vagrant, Chef cookbooks., MongoDB

* Teaching a lab session every week on a programming course Scheme.* Interacting with students and solving their doubts during office hours.* Correcting midterm papers, final's papers and weekly assignments.

I specifically wrote code to perform an end to end operation on a new reordering feature of the dimensions between the rows and the columns of the Pivot table structure, I was working on Adobe Flex, wrote service layer spring code and wrote hibernate queries to retrieve the right data. **I also worked on a collaborative dashboard interface for the product.