We are looking for an experienced software engineer with a performance oriented view of coding and knowledge in parallel frameworks. Someone who likes to have an end-to-end understanding of HW-SW, and make an impact on the world of accelerated computing to join our SDK group.
Responsibilities:
Play a key role in defining the architecture for the parallel compute framework API, and how it will interact with other common parallel compute frameworks. Integrate parallel frameworks using our SDK.
Ownership and accountability for accelerated solutions of common math libraries using NextSilicon SDK. Owns features end-to-end from requirements to testing.
Play a key role in defining and implementing NextSilicon SDKs and APIs.
Implement accelerated solutions and innovate by understanding our full software stack end-to-end. Identify performance bottlenecks and resolve them.
Work in close collaboration with cross-functional and multidisciplinary teams, including software, hardware, system, research, and apps engineering teams.
Requirements:
Develop and optimize algorithms for high performance execution.
5+ years of advanced C or C++ experience, with excellent coding skills.
MSc/BSc in CS\EE or other equivalent experience.
Hands-on experience working with common math accelerated libraries: BLAS, FFTW, Eigen, LAPACK.
Experience in parallel compute frameworks (such as OpenMP, MPI, CUDA, OpenCL, Kokkos) – big advantage.
Experience with compilers (familiarity with LLVM-IR, MLIR or proprietary) – advantage.
Develop and optimize algorithms for high performance execution.
5+ years of advanced C or C++ experience, with excellent coding skills.
MSc/BSc in CS\EE or other equivalent experience.
Hands-on experience working with common math accelerated libraries: BLAS, FFTW, Eigen, LAPACK.
Experience in parallel compute frameworks (such as OpenMP, MPI, CUDA, OpenCL, Kokkos) – big advantage.
Experience with compilers (familiarity with LLVM-IR, MLIR or proprietary) – advantage.
This position is open to all candidates.
