Joe Eaton is the technical lead for Accelerated Graph and Data Analytics at NVIDIA. Previously, he was chief scientist at Object Reservoir, producing finite-element based 3phase reservoir simulations with automated adaptive meshing technology. He was also vice president of engineering at Algebraix Data, leading a development team to produce a high-performance SPARQL query engine. Joe joined NVIDIA in 2013 to work in the CUDA Libraries team on sparse linear algebra methods and especially the AmgX library of GPU-accelerated sparse iterative solvers. Joe has experience in multi-phase and multi-physics simulators on large parallel computers, including convection-diffusion-reaction systems with millions of cells and tens of millions of chemistry unknowns. Currently he is the architect and technical lead on nvGRAPH, which applies sparse linear algebra and machine learning techniques for graph theory problems. Joe lives in Austin, Texas, and holds a Ph.D. in computational and applied mathematics from UT Austin, a master's in mechanical engineering from Stanford, and a bachelor's in mechanical engineering from Rice University.