AI / Machine learning / Computer Vision

Matthew
Benningfield

AI/ML Engineer

GitHub

I am an AI and machine learning engineer with a background in mathematics and applied statistics.

My professional work spans predictive modeling, time-series forecasting, and natural language processing.

My research interests include computer vision, machine learning, representation learning, and statistical modeling, with particular interest in connecting mathematical and statistical methods with modern approaches to intelligent systems.

01 / Areas of interest

Research interests

My interests center on how intelligent systems learn representations, interpret visual information, and operate in complex environments.

  • Computer vision & computational perception
  • Machine learning & representation learning
  • Multimodal learning & vision-language models
  • Mathematical & statistical modeling

02 / Applied work

Professional
experience

AI/ML Engineer III

May 2026 — Present

Aya Healthcare San Diego, CA · Remote

  • Developed automated data preparation and validation workflows for a Workforce AI scheduling system using historical and current time-series data.
  • Designed processes to improve data quality and provide reliable inputs to production predictive models.

Data Scientist II

April 2021 — May 2026

Humana Louisville, KY · Remote

  • Developed a national Nurse Practitioner availability forecasting model using Holt-Winters Exponential Smoothing across more than 50,000 providers.
  • Built an XGBoost regression model using Python, PySpark, and Databricks to predict inpatient admissions.
  • Developed a LightGBM authorization auto-approval system achieving 99% accuracy across approximately 385,000 monthly authorizations.
  • Used OASIS assessment data and XGBoost to forecast inpatient transfers and assign risk scores to more than 350,000 home health patients.
  • Automated compliance reporting for 1.9 million patient health metrics using Power BI.

Data Scientist

August 2020 — April 2021

True Rx Management Services Washington, IN · Remote

  • Designed ETL pipelines using Azure Data Factory, SQL, and Python to process pharmacy claims data.
  • Migrated Excel-based reporting to SQL Server and implemented more than 10 automated data pipelines using Azure DevOps.
  • Automated data-processing workflows to improve analytical reliability.

Business Intelligence Application Developer

October 2018 — August 2020

Kimball Electronics Jasper, IN

  • Developed and deployed an NLP sentiment analysis application for human resources.
  • Managed more than 10 global data pipelines across multiple ERP systems.
  • Automated audit compliance reporting with Python and Excel, reducing external audit costs by more than $50,000 annually.
  • Administered BI platform security across Power BI, SQL, and SAP BusinessObjects for more than 100 global employees.

03 / Academic background

Education

M.S. in Applied Statistics

December 2019

Rochester Institute of Technology Rochester, NY

B.S. in Mathematics

May 2015

University of Southern Indiana Evansville, IN

04 / Methods & tools

Technical expertise

Machine learning & AI
Supervised learning, predictive modeling, deep learning, natural language processing, large language models
Mathematics & statistics
Mathematical and statistical modeling, probability, time-series analysis, forecasting, regression, statistical testing
Programming
Python, C++, SQL, SAS, NumPy, Pandas
ML frameworks
PyTorch, TensorFlow, Scikit-Learn, XGBoost, LightGBM
Data & distributed computing
Apache Spark, PySpark, Databricks, Snowflake, ETL/ELT
Engineering & infrastructure
MLflow, Git, GitHub, Microsoft Azure, AWS, Azure DevOps, Linux

05 / Get in touch

Contact

mattbenningfield [at] gmail [dot] com

github.com/matbenni