I work for NVIDIA on parallel programming models (Fortran, C++, MPI, PGAS) for GPU and ARM platforms.
From 2014 to 2021, I worked for Intel on high-performance computing projects, including oneAPI and SYCL/DPC++.
From 2009 to 2014, I worked at Argonne National Laboratory, where I focused on application and system software for IBM Blue Gene supercomputers and contributed to the development of MPI-3 and its ecosystem.
Before I got into parallel computing full-time, I was a theoretical and computational chemist. I've been a developer of NWChem since 2006.
My website has details.
- Stack Overflow 4.7k 4.7k 33 gold badges2222 silver badges4343 bronze badges
- Computational Science 1.9k 1.9k 1414 silver badges1818 bronze badges
- Travel 451 451 33 silver badges77 bronze badges
- Matter Modeling 300 300 11 silver badge99 bronze badges
- Chemistry 201 201 44 silver badges99 bronze badges
- View network profile
Top network posts
- 165 MPICH vs OpenMPI
- 35 Why are airline passengers asked to lift up window shades during takeoff and landing?
- 32 Why use _mm_malloc? (as opposed to _aligned_malloc, alligned_alloc, or posix_memalign)
- 30 What features do users need from an MPI C++ interface?
- 21 Is half precision supported by modern architecture?
- 14 Can OpenMP be used for GPUs?
- 10 What is the largest system that has been studied by the coupled cluster method?
- View more network posts →