I have deep and broad experience in tackling challenging machine learning problems, designing systems to solve them, implementing those systems using a wide array of software technologies, and delivering the final products to users.
See the demo page for examples of systems I've built.
Designed and built integrated disease risk models using MRI images, genomic data and other information
using Keras, TensorFlow, VTK and ImageJ.
Built predictive models for phenotypes.
Put together an RNA pipeline from existing tools and some custom code for gene expression, allele specific expression, gene fusions and skipped exons in tumor tissue.
Used information theoretic tools to improve the performance of microbiome models.
I taught Computational Biology with Java. I published the course material as the book Mastering Java through Biology.
Designed and built the trainers and recognition components of MoboMath, a handwriting recognition system for mathematical notation. See the MoboMath page for more details.
Designed and implemented an HMM based cursive handwriting recognition system, including
an innovative distributed trainer.
Helped design and implement a Chinese handwriting recognition system.
Improved a biometric signature verification system.
Designed and built ydb, an innovative symbolic debugger for yacc compatible grammars and parsers. The system was purchased by UC Berkeley, IBM, and others.
Contributed to a number of projects in the functional languages group.
Built a speech compression system.
University of Wisconsin
Thesis: Cohen-Macaulay Complexes and Group Actions
University of Wisconsin
Oberlin College
Highest honors
