Christopher Krapu

bay area / north dakota

Currently working on internal AI tools @ NVIDIA.

I received my graduate training in statistics and engineering at Duke University under Mark Borsuk, graduating in 2020. I was awarded full-funding fellowships by NASA (NESSF, 2017) and the NSF (IGERT, 2014) with additional support from Amazon and NVIDIA. My research focused on developing graphical models for high-dimensional environmental data, with applications in engineering and spatial statistics.

latest posts

Apr 13, 2025	Modeling data with correlated errors across a directed graph
Jan 15, 2025	To my junior collaborators, this is how I want you to write your research code
Sep 28, 2024	Solving climate change by abusing thermodynamic scaling laws
Nov 20, 2022	Modeling temporal data with an unknown number of changepoints
Mar 07, 2022	Distributed zonal averages for fast geospatial analyses
Nov 28, 2021	Fast Kronecker matrix-vector product with einsum
Nov 10, 2021	Balanced spatial partitioning for point data in 20 lines
Apr 29, 2021	Modeling spatial structure in binary data with an H3 hexagonal coordinate system
Apr 05, 2021	Surrogate modeling for SEIR dynamics
Mar 30, 2020	Density estimation for geospatial imagery using autoregressive neural models
Feb 18, 2020	Multivariate sample size for Markov chains