Kyle enjoys using data science, remote sensing, and spatial statistics to analyze big, messy data and complex environmental and social systems.
Kyle has a background in environmental science and engineering, with a specific focus on geochemical techniques applied to water. His previous work includes developing sediment contaminant chronologies for the Hudson River at Rensselaer Polytechnic Institute (Troy, NY), and work in developing low-cost filters with arsenic adsorbents at Clarkson University (Potsdam, NY). His most recent graduate work at Tufts University used agent-based models to investigate the role of social and behavioral factors on the feasibility of point of use water technologies.
He co-taught various courses at Harvard Extension since 2014. In 2016, he worked as a GIS Analyst and later a Statistics Specialist for Tufts, providing individualized consultations and statistical services to students and faculty. He is a trained Software Carpentry instructor and a founding member of NESCLiC, and cares deeply about making statistics and data science methods understandable and accessible to all.
He is currently the Senior Data Science Specialist at Tufts University, providing statistical consulting, data visualization and high-performance computing (HPC) support.
Senior Data Science Specialist
Continue to provide advanced data science consultations. Developed analysis dashboard for partners including the World Bank using Tableau and RShiny. Co-designed the new Data Analytics master’s program and generated lecture materials on machine learning, regression models, and scalable computing.
2017 - 2018
Statistics and Research Technology Specialist
Provide individualized statistical consulting services to students, staff, and faculty, develop workshops and tutorials, and bridge high-performance computing with GIS and Data Lab services.
2016 - 2017
Provided GIS services to students, staff, and faculty, and implemented acquisition and storage of new spatial datasets.
Tested ceramic water filters, generated novel computational models for water flow through filters, developing agent-based models predicting the influence of behavioral factors of water treatment. Over 100+ hours of teaching experience.
Developed a novel ceramic water filter to remove arsenic, removal of organics from groundwater with ISCO, encapsulation of adsorbents for increased stability and timed release. Over 200+ hours of teaching experience.
Foraminiferal Analysis Technician
Recovered core samples from Hudson River sediment cores and processed through weighing and wet sieving, along with foraminifera fossil recovery and classification.
Assisted students with statistical and geographic model development, troubleshooting and analysis.
2016 - 2017
Data Lab Assistant
Further focused my training on developing statistical models of health interventions. Gained intermediate skills in statistical analysis and geographic information system (GIS).
ENVIRONMENTAL SCIENCE & ENGINEERING
Gained extensive field sampling and lab testing experience for water and air samples, developed an arsenic testing methodology and organized the Clarkson Ceramics Remediation Group (CCERG).
ENVIRONMENTAL SCIENCE & PSYCHOLOGY
Worked on sedimentary analysis for contaminant detection and fossil recovery for climatic reconstruction.
Machine learning & advanced statistical models
Dashboards (Tableau, Python Flask, RShiny)
Traditional web application deployment (Tomcat, SSL)
Geographical Information Systems (GIS) - QGIS & ArcMap
Statistics (Stata, SPSS, SAS, R)
Big data analysis (PySpark, AWS, SQL)
Wherever the machine is, I can develop a dashboard and deploy a model onto it. I have experience with deploying analyses over ssh to a bare Linux box, AWS, Azure, Windows Server, and local installs. Making code effective, correct and efficient is a joy to me. Once I learn insights about the data, communicating those insights clearly to my team is paramount to me.
Today we live in a world of "big data" and be need to carefully use techniques to extract conclusions and take actions. I can use statistical software (Stata, SPSS, R) to investigate and visualize your data set. I have gone through substantial statistical training and you can find examples of my work on my GitHub.
Monitoring contaminants in air and water is complex. I am versed in air and water contaminant detection including:: mass spec, UV-Vis spec, ICP-AES, XRF, Flame-AA,
SEM, GC-FID/ECD, MIRAN, GRIMM, CPC, and numerous other air sampling methods.
I am also comfortable working with this data to extract conclusions, especially around spatial clusters and time series.