Senior Applied Research Scientist – GPU Native Numerical Algorithms
Core
Design GPU-native numerical methods and solver algorithms to accelerate engineering simulations and enable AI-native computational methods.
Role type
Senior Applied Research Scientist (GPU Numerical Algorithms)
Builds
GPU-native solver prototypes and research platforms for engineering simulation
Domain
Accelerated computing, scientific computing, computational engineering
Deliverable
production ML models
Required skills
GPU-native algorithm design, linear/nonlinear solvers, sparse linear algebra, C++, CUDA, performance profiling, PDE discretization
Preferred skills
Implicit structural dynamics, nonlinear mechanics, CFD, electromagnetics, multiphysics, distributed solvers (MPI/NCCL), PETSc, Trilinos, NVIDIA Warp/CUDA-X
Technologies
CUDA, C++, Python, MPI, NCCL, cuSPARSE, cuSOLVER, PETSc, Trilinos, MFEM, libCEED, OpenFOAM
Responsibilities
Invent and reformulate numerical algorithms co-designed for NVIDIA GPU architectures; Develop linear and nonlinear solver approaches; Evaluate algorithms on workloads in mechanics, thermal-fluid, electromagnetics, and semiconductor simulation; Collaborate with engineering teams to move research prototypes to software capabilities; Shape the long-term applied research roadmap for GPU-native numerical methods.
Seniority
Senior, hands-on IC
