● Created configurable Pyspark ETL framework for large drug company | Python, PySpark, SQL, Azure, ADF, Databricks

● Designed and implemented Covid Data Parser; extended ETL framework with the new features; coached junior engineers to enable data science use cases for fitness studio franchise | Python, PySpark, Athena, Glue, Assemble, Terraform, AWS, S3

● Implemented raw data ingestion and transformation from different sources and provided continuous support to data scientist team for multi-brand restaurant company | Talend, ADF, ADLS, SQL Server, Databricks, OpenWeather, SportRadar, PySpark

● Implemented data transformation pipelines and built application to get information on connected components to classify the customers for giant movie theater chain | AWS, EMR, PySpark, Python, Dash, DocumentDB

● Deployed ML models on Azure (data preparation, build models, docker custom image build) | Python, Azure ML Workbench, Azure ML Studio, docker, pickle, sklearn, pandas, numpy, kubernetes, Azure Function

● Implemented data lake and end-to-end data flow on Azure, generated reports on Tableau from the final tables | Data Factory, Azure SQL Server, Azure Blob, SQL Stored Procedures, SQL, JSON

● Built data preprocessing pipeline using Amazon Redshift and S3, worked on feature engineering,

descriptive statistics, statistical testing and visualizations as a part of a project to predict demand for one of the

largest brewing companies in the world | Python, SQL, pandas, numpy, scipy

● Developed front-end and back-end functionality, worked on data quality assessment, cleaning, analytics and

visualization of a dashboard to surface relevant fleet data to account managers for a Fortune 500 transportation

services company | Javascript, React, Node.js, Python, SQL, pandas, numpy, matplotlib, graphlab

● Successfully implemented a number of machine learning models from first principles | Python, MATLAB, Math

Responsible for the infrastructure required to stage the results of a large computational grid used to determine capital reserves required to meet federal Basel 3 requirements:

● Implemented the path aggregation service used to collect, normalize and store the price results and diagnostic information. The aggregator is a multi-threaded C++ application which consists of central processing server and a client library | C++, ZeroMQ, Protobuf

● Wrote C++ and Python libraries for serialization of data and objects using the HDF5 library

● Responsible for managing the build process for the pricing, simulation, and aggregation services. This required configuration of a Jenkins continuous build environment, CMAKE configuration files supporting Linux and Windows builds, and implementation of Python control scripts | Python, CMake

Worked alongside the CTO and head developer to test new features across web browsers for www.thefuturefm.com. Provided day-to-day technology support, uploaded web content (i.e. new music file uploads and DJ profiles) and managed a proprietary MixScan software for digital music assets. Maintained a quality assurance database and provided trouble-shooting, quality assurance, testing and user experience feedback.