Here are some of my projects and publications. You can find more coding projects on my Github profile.
I'm the cofounder and CTO of a transportation analytics startup called Percepsense. Our aim is to bring the best of the 21st century: crowdsourced analytics, cutting edge machine learning,
and the ubiquity of smartphones, into an industry where it is desperately needed.
Our team graduated from CMU's NSF-ICorps 2017 Fall Cohort.
We are currently in the process of partnering with Honda Automotives.
We built an iPhone app that detects and plots potholes from smartphone sensor data. Read about it Here
Paper accepted into DSCO 2017.
realtime pothole detection and visualization. Includes iPhone app source, iPython Notebooks, and trained weights.
Non-deterministic, iterative variant of PCA designed to work on huge, noisy datasets with missing and imperfect data. Implemented in PyTorch for trivial GPU accelerated dimension-reduction.
Selected keynote presenter
Data Science For Social Good Conference - Selected speaker