Developer Technology Engineer, Energy

Our work at NVIDIA is dedicated towards a computing model focused on visual and AI computing. For two decades, NVIDIA has pioneered visual computing, the art and science of computer graphics, with our invention of the GPU. The GPU has also shown to be spectacularly effective at solving some of the most complex problems in computer science. Today, NVIDIA’s GPU simulates human intelligence, running deep learning algorithms and acting as the brain of computers, robots and self-driving cars that can perceive and understand the world. We are looking to grow our company and teams with the smartest people in the world and there has never been a more exciting time to join our team!  NVIDIA   is   looking   for   a   passionate, world-class computer scientists and engineers ( Compute   Developer   Technology  - DevTech)   to   accelerate   Energy   simulation   and   AI   workflows   on   NVIDIA   platforms.   You   will   focus   on   CUDA   performance   optimization   for   workloads   such   as   seismic   processing   (e.g.,   imaging/inversion   pipelines),   reservoir   simulation ,   power   grid   simulators ,   and   related   HPC/AI   production   workflows.   You   will   work   hands-on   with   customer   and   partner   engineering   teams   as   well   as   NVIDIA   product   and   engineering   groups   to   deliver   measurable   speedups   and   scalable   performance   on   multi-GPU   and   multi-node   systems. What  y ou will  b e  d oing: Profile,   analyze,   and   optimize   GPU-accelerated   applications   with   emphasis   on   CUDA   kernels,   memory   movement,   concurrency,   and   end-to-end   throughput . Drive   performance   improvements   across   the   stack: CUDA   C++   kernel   optimization,   launch   configuration,   memory   hierarchy,   streams/events GPU   libraries   (as   applicable):   cuBLAS,   cuFFT,   cuSPARSE,   cuSOLVER,   NCCL Multi-GPU   and   multi-node   scaling   using   MPI   +   NCCL ,   CPU/GPU   overlap,   communication   patterns Build   reproducible   benchmarks ,   performance   reports,   and   tuning   recommendations   (before/after,   methodology,   scaling   curves). Develop   and   maintain   reference   implementations,   examples,   and/or   patches   to   customer   code   to   enable   performance   and   portability. Support   customer   engagements   (POCs   to   production),   including   debugging   correctness/performance   issues   and   advising   on   best   practices   for   deployment   (containers,   schedulers,   clusters). Collaborate   with   internal   teams   to   file   actionable   issues,   validate   fixes,   and   influence   roadmap   based   on   real   customer   requirements   in   Energy. Build internal libraries and resusable code that would lead to future NVIDIA products. What  w e  n eed   to  s ee: BS/MS   (or   equivalent   experience)   in   CS/CE/EE/Physics/Applied   Math   or   related   field. Strong   programming   skills   in   C/C++   and   Python   on   Linux. Hands-on   experience   with   CUDA   programming   and   GPU   performance   optimization   concepts. Experience   profiling   and   debugging   performance   using   tools   such   as   NVIDIA   Nsight   Systems   /   Nsight   Compute   (or   equivalent). Understanding   of   parallel   computing   and   performance   fundamentals   (vectorization,   threading,   NUMA,   memory   bandwidth/latency). Ability   to   communicate   technical   findings   clearly   to   both   engineers   and   non-engineers. 5+   years   relevant experi ence   in   GPU/HPC   optimization;   strong   track   record   of   delivered   speedups   and   scaling   improvements. Ways to stand out