David Smith Email & Phone Number

Boston, Ma, Us

Boston, Ma, Us

Data Engineer (May 2019 - March 2020) | Senior Data Engineer (March 2020 - present)- Awarded 2019 Manulife Star of Excellence, US Segment- Designed and deployed API for reading and writing to enterprise data lake tables stored in HBASE and Hive- Implemented Prometheus and Grafana monitoring for spark batch jobs and APIs hosted by the data lake.- Delivered data ingestion projects from origin sources such as DB2, SQLServer, and Salesforce to the enterprise data lake- Coordinated support documentation and knowledge transfer for extended offshore support teams and transitioning maintenance of developed applications to other teams

- Developed predictive random forest models to predict historic discharge disposition of patients based off of claim histories using scikit-learn and python- Provided custom analytics and data characterization through python and pandas of MassHealth Medicaid patient enrollment and claim files to customer hospitals- Worked on ETL pipelines for the loading of Excel spreadsheet reports into a SQL/Django backend.- Contributed features to the front-end of the web application using React + Redux with ES6 Javascript

Woods Hole, Massachusetts, Us

- Co-authored research in Nature Partner Journal | Climate and Atmospheric Science: see https://www.nature.com/articles/s41612-020-0120-6- Designed and supervised automatic downloading and updating of over a terabyte of climate data from multiple sources using Python and cron- Analyzed atmospheric blocking in different regions of the North Atlantic Ocean and the skill of current climate models in predicting the frequency and variability of blocking events over 55 years using MATLAB- Evaluated relative predictive skill of new climate model with 4 times more weighted starting scenarios against existing models to determine whether enough starting scenarios were considered

Boston, Ma, Us

Worked as member of the Research and Development team, with a focus on writing optimization pipelines and ad-hoc analytics to support ad campaigns. I wrote more than 10 Python pipelines to automatically optimize real time bidding for ad campaigns in python, and led restructuring of core budgeting algorithms to increase time performance. I also performed ad-hoc data analytics from SQL tables and data in Apache HBase according to business needs and campaign performance. Finally, I developed a new algorithm to rapidly detect and halt the overspending of ad campaign budgets, leading to an increase in net margin by 3%.

Greenbelt, Md, Us

Worked on parallelizing code that is very computationally expensive to run (maximum of around 5 days) to reduce runtime. This involved researching parallelization techniques in IDL and C++, as well as selectively reading in a formatted database to perform operations on. In addition, I attended weekly meetings with other interns to discuss progress on projects and obligations of being an intern at NASA Goddard Space Flight Center.