Bob studies modeling, analysis, and optimization techniques for high-performance software, with a current focus on deep learning. His philosophy is that the combination of statistical methods, code analysis, and domain knowledge leads to better tools for understanding and building fast systems. Bob earned his B.S. at Northwestern in 2005, spent four years writing performance analysis tools for supercomputers at the DoD, and another three years at PNNL leading research in massively multithreaded architectures and high-performance graph analysis.