Experience

Harvard University

• United States
• Higher Education
• 700 & Above Employee

• Senior Software Engineer

  • Jan 2023 - Present

  Modernizing application architecture and database design to enable public access to over 100 years (and 1.5PB) of historic astronomical glass plate images. - Collaborating with stakeholders to define and refine product requirements and implement an agile development pattern - Designing and Implementing AWS based infrastructure using Terraform to replace an on prem system and enable greater resilience and scalability and ensure preservation of this valuable data far in to the future - Developing a graph based data model to enable discovery of new historic connections and allow users from all backgrounds to explore and experience the history of astronomy - Designing and implementing RESTful APIs using Go and Python for secure data access while also working to minimize cloud costs Show less



Roivant Sciences

• United States
• Biotechnology
• 100 - 200 Employee

• Senior HPC Applications Engineer

  • Sep 2021 - Nov 2022

  Scaled up scientific computational workloads to take advantage of Roivant’s on prem GPU HPC cluster and Google Cloud HPC infrastructure. Developed tooling and systems to monitor and improve overall system utilization. Collaborated with software engineering teams to architect distributed applications and ensure resources were used efficiently. Specific projects included: · Deployed molecular docking workflow across 100s of instances using GCP batch. Increased efficiency to improve run-time from hours to minutes by streamlining administrative and computational overhead. · Designed distributed hybrid logging architecture utilizing Google logging, and fluentbit. Enabled unified logging in multi-process and multi-node applications with minimal overhead for developers. · Developed and deployed executive and user facing dashboard showing historic and real-time cluster usage data using Telegraf, InfluxDB, Prometheus, and Grafana. Enabled quick identification of system problems and clear project prioritization decisions. · Collaborated with data scientists and data engineers to deploy a containerized ML Flow instance for ML experimentation and model registration. Enabled organization wide access of ADME property prediction models. · Mentored scientists and other engineers in use of HPC and cloud parallelization technologies. Enabled junior engineers to run workloads at scale. · Supported junior engineers and scientists in taking advantage of CI/CD to automatically build containerized applications and serve them from a central registry. Improved scientific code bases through implementation of modern software best practices. Technologies Used: Singularity, Docker, GCP, Pytorch Lightning, Ray, Fluentbit, Slurm, Cuda, Ml Flow, Prometheus, Grafana, Telegraf, ZMQ, InfluxDB, Gitlab CICD, PostgreSQL, MySQL Show less



Harvard University

• United States
• Higher Education
• 700 & Above Employee

• Research Software Engineer

  • Jan 2021 - Aug 2021

  Provided software engineering and data science expertise to researchers and students as part of the research software engineering team at Harvard University. Developed data acquisition and cleaning pipelines for large nationwide datasets. Provided support and code optimization for researchers working with big data on HPC systems. Collaborated with researchers to understand and properly utilize CMS patient record data. Provided University Wide trainings on computational challenges facing researchers. · Developed an R package implementing a data preparation and Machine Learning pipeline for air pollution predictions. Simplified expression of modeling methods. Presented a paper at IEEE Data Science and Advanced Analytics 2018. · Developed R based data querying system for CMS patient data. Significantly improved turnaround time and reproducibility of datasets. · Developed Python packages automating data acquisition and data cleaning for key data sources as part of data platform. Enabled creation of complex data sets by researchers through use of simple configuration files. · Collaborated with researchers from multiple universities to utilize public APIs, GIS libraries, custom tools, and HPC resources to quickly create geospatial data sets for use in ML models and statistical analyses. Contributed to the publication of over 16 peer reviewed papers. · Developed materials for a training session educating researchers on working with large data sets in the R language. Enabled researchers without computational backgrounds to improve memory management of statistical programs. Technologies Used: R, Python, Slurm, CWL, GDAL, Conda, ggplot Show less



Harvard T.H. Chan School of Public Health

• Boston, Massachusetts

• Research Assistant III

  • Sep 2017 - Jan 2021

  I work on various data science applications within the Department of Biostatistics developing code to implement statistical models examining air pollution and its connections to health. I also work on setting up data sources for use with cluster computing systems, as well as leveraging those systems to assist with the statistical analysis. Most projects involve processing data much larger than the memory of a standard personal computer. Specific Projects have included the development of an R package facilitating the use of machine learning algorithms for environmental health researchers and the design and implementation of database systems to ease the creation of analytic datasets. Show less



Institute for Quantitative Social Science (IQSS)

• United States
• Higher Education
• 1 - 100 Employee

• Data Science Intern

  • Jun 2017 - Sep 2017

  I Assist with the development, documentation, and maintenance of open source statistical software packages released by IQSS. Specific projects include work involving designing visualizations for complex statistical calculations, work with time series models, work involving neural networks, and development of automated checks for the released packages. I Assist with the development, documentation, and maintenance of open source statistical software packages released by IQSS. Specific projects include work involving designing visualizations for complex statistical calculations, work with time series models, work involving neural networks, and development of automated checks for the released packages.



Epic

• United States
• Software Development
• 700 & Above Employee

• Technical Services

  • Aug 2016 - May 2017

  I provide support to customers using Epic's electronic health record ETL infrastructure. I help troubleshoot errors in data content, as well as aiding in the transfer of data from non-relational to relational structures. In addition to these responsibilities, I leverage my background in data analytics to help provide statistical insights for the team. I provide support to customers using Epic's electronic health record ETL infrastructure. I help troubleshoot errors in data content, as well as aiding in the transfer of data from non-relational to relational structures. In addition to these responsibilities, I leverage my background in data analytics to help provide statistical insights for the team.



Boston University

• United States
• Higher Education
• 700 & Above Employee

• Academic Tutor

  • May 2014 - May 2016

  I provide individual and group instruction to student-athletes from a variety of academic backgrounds on a number of economic concepts including mathematical modeling of economic relationships, the nature of economic growth and development, and statistics. In addition to instruction, my duties include independent scheduling of tutoring sessions and working with administrators and coaches to best help the students.

• Research Assistant

  • May 2013 - May 2014



Boston University Economics Department

• Research Assistant

  • May 2015 - Aug 2015

  Wrote scripts in Python, R, and Stata to automate research processes and analyses for Dr. Francesco Decarolis. Successfully created a program to rapidly process data from distributed web sources and assemble it into a format for easy analysis. Additionally wrote code for reformatting big data sets and initial pattern analysis within the data. Wrote scripts in Python, R, and Stata to automate research processes and analyses for Dr. Francesco Decarolis. Successfully created a program to rapidly process data from distributed web sources and assemble it into a format for easy analysis. Additionally wrote code for reformatting big data sets and initial pattern analysis within the data.



The Borgen Project

• United States
• Non-profit Organizations
• 400 - 500 Employee

• Political Affairs Intern

  • Sep 2014 - Jan 2015

  I was responsible for contacting congressional leaders in support of US foreign aid, raising awareness about global poverty issues and legislation, as well as fundraising. I successfully surpassed all fundraising goals, as well as increasing awareness of the Borgen Project and its opportunities among members of the service oriented community at Boston University. I was responsible for contacting congressional leaders in support of US foreign aid, raising awareness about global poverty issues and legislation, as well as fundraising. I successfully surpassed all fundraising goals, as well as increasing awareness of the Borgen Project and its opportunities among members of the service oriented community at Boston University.



Let's Get Ready

• Boston

• Head Verbal Coach

  • 2013 - 2014

  Worked as a group instructor for a class of 7-8 students from underserved communities in the greater Boston area. I taught skills required for the verbal section of the SAT while also aiding the students in the writing process for their Common Application personal statements, resulting in improvements in SAT scores and completed rough draft essays. My second summer I was selected to work as the head coach for my location. Additional duties included the creation of model lesson plans and providing support and guidance for the other teachers. Show less