Remote CUDA HPC Engineer – GPU & ML Optimization

United States Digital Space LLC; United States; Remote; USD 100,000 - 150,000; Full time

Job summary

Bright Vision Technologies seeks an experienced Parallel Computing Engineer to optimize AI, ML, and HPC workloads on GPUs. You will design high-performance CUDA kernels, implement distributed training, and improve inference throughput on large models.

The role is fully remote within the United States, offering strong growth potential and exposure to cutting-edge GPU libraries and frameworks. If you have 6+ years in GPU programming and a passion for performance engineering, join us.

Qualifications

• 6+ years of GPU programming and performance optimization.
• Strong CUDA C/C++ expertise and parallel programming.
• Experience with NCCL and MPI in high-performance networking.

Responsibilities

• Design and optimize CUDA kernels for AI and HPC workloads.
• Profile and tune GPU applications with Nsight tools.
• Optimize memory, multi-GPU performance, and distributed training with NCCL and MPI.
• Develop operators and kernels for PyTorch, JAX, or Triton.
• Improve training and inference performance for large-scale ML workloads.
• Build benchmarks and document best practices.
• Collaborate with ML/engineering teams to deliver scalable GPU solutions.

Skills

GPU programming

CUDA C/C++

Parallel programming

C++ debugging

Distributed computing

Education

Bachelor's/Master's degree in CS/CE

Tools

NCCL

MPI

Job description

Bright Vision Technologies seeks an experienced Parallel Computing Engineer to optimize AI, ML, and HPC workloads on GPUs. You will design high-performance CUDA kernels, implement distributed training, and improve inference throughput on large models.

The role is fully remote within the United States, offering strong growth potential and exposure to cutting-edge GPU libraries and frameworks. If you have 6+ years in GPU programming and a passion for performance engineering, join us.