Josh Arnold Email & Phone Number

Herndon, Virginia, United States

Served as Spark and Scala expert for the Measures Team, responsible for computing quality-of-care metrics for the Centers for Medicare & Medicaid Services public-facing quality metrics reporting services. By targeting Scala-Spark optimizations and fine-tuning Spark configurations, reduced memory footprint of production jobs by as much as 85%.

Served as data engineer for the REACHVET project. Architected and implemented a Spark-HDFS solution for ETL feature-extraction pipelines, leading to 2500% speed-up over existing approach. Designed and tested an entity-centric Elasticsearch data-model for improving human-in-the-loop feature selection processes. Also served as Co-PI for Data Science for the CMS Center for Program Integrity project. Designed on-premises data lake for tens of terabytes of data. Applied gradient boosted trees, transformer-based models, and SHAPley factor analysis to provide novel insights regarding waste, fraud, and abuse.Created data visualizations and presentations for communicating findings to sponsor and mentored junior data scientists.

Oak Ridge, Tn

Served as data engineer in collaboration with the Department of Veterans Affairs for the evaluation of state-of-the-art open-source technologies within the context of VA research and production workflows. Provisioned, deployed, tested, and benchmarked a myriad of open source storage and NLP analytics solutions.

Greater Nashville Area, Tn

Served as ML and big data expert for the university, including delivering lectures and seminars, creating demonstration projects, and collaborating on research as staff member of the Advanced Computing Center for Research and Education (ACCRE). Created "context_bot" for automating user ticket triage and designed the new ACCRE user database.

Gaithersburg, Md

Developed image classification techniques for hyper-dimensional microscopy. Simulated cement microstructure evolution using lattice-based cellular automata on a high performance cluster. Developed parallelized nanoscale models of cement microstructure utilizing CUDA texture memory.

Greater Nashville Area, Tn

Greater Nashville Area, Tn

Dissertation: Reducing uncertainty in the characterization and modeling of blended cement mortar

Petten, Netherlands

Aiken, Sc

Doctor Of Philosophy (Ph.D.), Environmental Engineering

Data Science

Master Of Science (M.S.), Environmental Engineering

Activities and Societies: Vice President of Academic Affairs

Bachelor Of Science (B.S.), Civil Engineering