Senior Engineer, Math Libraries - Consumer and Embedded Platforms
Core
Enable NVIDIA's CUDA Math Libraries (cuBLAS, cuSPARSE, cuSOLVER, cuFFT) on new and specialized platforms, ensuring functional correctness, performance readiness, and platform integration.
Role type
Senior IC software engineer (high-performance computing & GPU libraries)
Builds
CUDA Math Libraries for AI, data analytics, and scientific simulations
Domain
High-performance computing, GPU acceleration, numerical computing
Deliverable
production ML models
Required skills
C++, Python, parallel computing, numerical methods, HPC software architecture, performance optimization, CI/CD pipelines
Preferred skills
CUDA, GPU-accelerated computing, MPI, OpenMP, OpenACC, pthreads, BLAS, LAPACK, FFT, sparse solvers, embedded/safety-critical software development
Technologies
CUDA, cuBLAS, cuSPARSE, cuSOLVER, cuFFT, JIRA
Responsibilities
Ensure platform readiness for Math Libraries across new and specialized platforms; Perform defect triage and isolation of platform-specific versus library-specific problems; Identify performance targets and fix performance regressions; Collaborate with CI/CD and infrastructure teams to establish qualification and validation pipelines.
Seniority
Senior, hands-on IC
