Joe.
I build software to help humans thrive.
I believe that software can be used to improve the state of the world despite it’s challenges and potential dangers. I have a passion for developing applications that make the world a better place.
Hi, my name is
About Me
I believe machine learning can be used to improve the state of the world despite its potential dangers. This belief guides my approach to building. Good software goes a step beyond solving a problem, it should enable humans to live a better life.
I am currently a Senior Machine Learning Software Engineer at One Medical where I am working on automating labourous and monotonous tasks to give time back to health care providers so that they may better connect with their patients.
Here are a few technologies I've been working with recently:- Python
- AWS SageMaker
- Docker
- AWS Lambda
- FastAPI
- AWS ECS
Experience
I am currently working as a software engineer at One Medical where I am building machine learning powered applications to support an Electronic Health Record.
I researched and developed novel approaches to machine learning for behavioral change AI. I designed and developed an in-house experimenation platform as well as implemented core components of the training pipelines for a deep learning model.
- This repository contains the source code reproducing the numerical examples presented in the paper An adaptive stochastic gradient-free approach for high-dimensional blackbox optimization.
- Here is the official reopsitory for the paper ‘‘Offline Policy Comparison under Limited Historical Agent-Environment Interactions’’ that is available at https://arxiv.org/abs/2106.03934.
As a research assistant I designed, analyzed and implemented novel algorithms using approximation theory to solve a variety of problems in application domains including
- image processing
- signal denoising
- classification
More details can be found in my Ph.D. dissertation https://trace.tennessee.edu/utk_graddiss/5916/
Education
Recent Work
Python
Tesseract
Terraform
AWS Lambda
AWS SNS/SQS
Automated routing of faxed documents
Used Computer vision and deep learning to extract important information from incoming faxes. This service also automatically routes faxes to their correct final destination in an electronic health record (EHR) when the type is detected.
Python
Reinforcement Learning
Limited Data Estimator
Comparison of policies for contextual bandit problems using a set of limited historical interactions.
Python
Parallel processing (mpi4py)
Adaptive Stochastic gradient-free (ASGF) blackbox optimization
A black-box optimization method for extremely high dimensional non-convex objective functions.
Get in Touch
If you would like to contact me, feel free to email me. I can not guarentee an expedient response, but I’ll do my best to reply.