Kyle Heuton
Verified Expert in Engineering
Machine Learning Developer
Kyle is a scientific software developer with eight years of experience building data engineering applications for healthcare. In global health, he built mortality forecasting tools to forecast global deaths for a public health research institute. In US healthcare, Kyle built a data ingestion platform to receive and normalize data on hundreds of millions of patients.
Portfolio
Experience
Availability
Preferred Environment
Amazon Web Services (AWS), NumPy, Pandas, Scala, Python
The most amazing...
...project I've completed was building mortality forecasting tools to forecast deaths due to 205 different causes in 195 countries.
Work Experience
Software Engineer
OM1
- Designed a Python platform on the AWS cloud to receive, de-identify, and normalize electronic medical records and insurance claims data from diverse sources on hundreds of millions of patients.
- Deployed the ingestion platform to receive data from 10 different data partners to ingest GBs of data daily. This project was accomplished within one year.
- Constructed a data processing service in Scala to define cohorts of patients based on clinical disease criteria and enrich those cohorts with predictive metrics on disease outcomes and medical expenditures from machine learning models.
- Managed the data team, as the interim team lead, to develop data transmission procedures with customers. I also planned the team’s roadmap and mentored junior engineers.
- Created SQL queries and workflows to manage complex ETL tasks on medical data including de-duplication, patient linking, and deriving clinically relevant metrics such as insurance histories and drug eras.
Software Engineer and Forecasting Researcher
Institute for Health Metrics and Evaluation
- Built a predictive modeling platform to generate forecasts of health scenarios worldwide and the potential impacts of specific policies on global health as the team’s lead Spark engineer. Forecasted mortality from 205 causes in 195 countries.
- Developed a scientific software pipeline in Python used by dozens of modelers to run more than 20,000 models annually. Data and results were stored in our SQL database, and models run on a Univa Grid-Engine cluster.
- Created Python tools to support data analysts and researchers in modeling disease prevalence and economic drivers of health.
Experience
Data Ingestion Platform for TBs of Medical Data
I built a system on AWS using S3 buckets for storage. A serverless Lambda script listened to the buckets for any incoming files, and when a file arrived it would log its receipt and process the files accordingly. This platform was successfully deployed to 10 different partners within its first year, and it was ingesting GBs of data every day.
Health Profiles for Every Country in the World
Education
Master of Public Health Degree in Quantitative Health Metrics
University of Washington - Seattle, WA
Bachelor of Science Degree in Chemical Engineering
University of Minnesota - Minneapolis, MN
Bachelor of Science Degree in Mathematics
University of Minnesota - Minneapolis, MN
Skills
Libraries/APIs
Pandas, Scikit-learn, NumPy, D3.js
Tools
STATA, MATLAB
Languages
Python, SQL, Scala, R
Platforms
Amazon Web Services (AWS), Amazon EC2
Storage
Amazon S3 (AWS S3)
Paradigms
Data Science
Other
Data Analysis, Machine Learning, Regression, Claims
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring