Remote CUDA GPU Software Engineer — HPC & AI Performance

RiseMe; Monroeville (Allegheny County); On-site; USD 100,000 - 175,000; Full time

Job summary

Bright Vision Technologies is seeking a GPU Software Engineer (CUDA) to design and optimize compute-intensive workloads on modern accelerators for AI training and HPC. This remote U.S. role emphasizes delivering high-performance kernels and scalable multi-GPU solutions.

Ideal candidates will have deep CUDA expertise, strong C++ skills, and a track record of profiling and optimizing production GPU workloads, with collaboration across product, design, and engineering teams.

Qualifications

• We require 10+ years in GPU programming and performance engineering.
• Deep expertise in CUDA C/C++ and GPU programming models.
• Strong understanding of modern GPU architectures and memory hierarchies.
• Hands-on experience profiling and optimizing GPU workloads in production.
• Experience integrating custom kernels into ML frameworks.
• Strong C++ skills and modern systems programming practices.

Responsibilities

• Design and implement high-performance CUDA kernels for AI and HPC workloads.
• Profile and optimize GPU code using Nsight Systems, Nsight Compute, and CUDA profilers.
• Tune memory access, occupancy, and shared memory for peak performance.
• Develop optimized libraries for linear algebra and ML primitives.
• Optimize multi-GPU and multi-node training using NCCL and high-speed networking.
• Implement custom operators and fused kernels in PyTorch, JAX, or Triton.
• Collaborate with ML engineers to identify bottlenecks in training and inference.
• Develop benchmarks and regression tests to safeguard performance over time.
• Evaluate new GPU architectures and advise on adoption.
• Contribute to compiler-level optimizations for tensor programs where appropriate.
• Document performance characteristics and tuning playbooks.

Skills

CUDA C/C++

GPU programming

Performance profiling

C++

MPI/NCCL

Numerical methods

Communication

Education

Bachelor’s or Master’s degree in CS/CE

Tools

Nsight Systems

Nsight Compute

Triton

CUTLASS

TensorRT

LLVM/MLIR

Job description

Bright Vision Technologies is seeking a GPU Software Engineer (CUDA) to design and optimize compute-intensive workloads on modern accelerators for AI training and HPC. This remote U.S. role emphasizes delivering high-performance kernels and scalable multi-GPU solutions.

Ideal candidates will have deep CUDA expertise, strong C++ skills, and a track record of profiling and optimizing production GPU workloads, with collaboration across product, design, and engineering teams.