[ E-Commerce Listing AI ]

Built an AI system with ChatGPT to automate writing product listing pages

[ Product Thumbnail Generation AI ]

Developed method to intelligently generate new image backgrounds for product listings

[ Fashion Product Classification AI ]

Built a zero shot visual system to identify clothing/accessories in an image and label their style, color, material, ect

As owner of the backend, I handle all new features, bug fixes, and admin tasks.

I also do all of our DevOps including our CI/CD pipeline, network configs, and serverless deployments.

[ 3rd Party Integration Services ]

[ File Ingestion Pipeline ]

[ ElasticSearch Upgrade/Migration ]

[ Handwritten Signature Matching ]

Trained deep learning model and wrote microservice to match handwritten signatures on document to their most likely author, achieved 90% accuracy.

Used Tensorflow, Python, and Rust

[ AI Construction Project Estimation Software ]

Solutions Architect on browser based project estimation and blueprint editing software.

Responsible for taking product from initial idea through development into production.

Developed around half of the front-end features, entire api + microservices, as well as details of UI design.

Used Node.js, React, Express, MongoDB, and Stratum.

[ 3D Medical Animation Tool ]

Architect on project to add animation capability to in-house medical 3D modelling software.

Led team of 4 developers, responsible for architectural design as well as front end development.

Used C++, OpenGL.

[ Drivers License Parser ]

Wrote a tool in 2 weeks that extracts name, address, ect. from drivers licenses.

Used Python, OpenCV, Numpy.

[ Deep Learning MRI Injury Detection System ]

Core developer on project for diagnosing injuries from raw MRI scans.

Handled system architecture, wrote initial research prototype, oversaw development, testing, and delivery of final product.

Managed machine learning development team.

Resulted in 2 peer reviewed publications, with 3 in review.

Currently shepherding product through FDA approval.

Used Python, Tensorflow/Keras, Numpy

[ Document Classification AI ]

Built automated document classification prototype. Used image and text features with gradient boosted forests, deep learning, and NLP techniques for classification, keyword extraction, and priority classification and LSTMs for handwriting recognition.

[ Semantic Text Extraction ]

Completely refactored text extraction system for medical reports; added regex extraction and k-means clustering for semantic parsing.

Designed and developed software tools for simulation and testing of IoT devices and wrote algorithms for 3D location/ranging. Python on Ubuntu Linux.

Developed a manufacturing software image with REST API for nerve ablation machine as well as bug fixes and features. Wind River Linux with C++ with Qt, Python for unit tests.

Wrote manufacturing tests with for a portable defibrillator with Python.