Remote CUDA Kernel Engineer - AI GPU Performance

Pragmatike; New Jersey; On-site; USD 110,000 - 150,000; Full time

Benefits offered by this job

Competitive salary & equity options

Sign-on bonus

Health, Dental, and Vision

401k

Job summary

Pragmatike is hiring a CUDA Kernel Engineer to develop and optimize NVIDIA CUDA kernels for a leading AI startup. This remote role focuses on maximizing GPU performance and throughput for high-scale AI systems. Candidates should have substantial experience with CUDA and a profound understanding of GPU architecture. The position includes competitive salary, equity, sign-on bonuses, and comprehensive health benefits. Join a fast-paced environment that’s shaping the future of AI.

Qualifications

• Proven track record building NVIDIA CUDA kernels from scratch.
• Strong ability to optimize kernels with advanced techniques.
• Deep understanding of GPU memory hierarchy and performance bottlenecks.

Responsibilities

• Design, implement, and optimize custom CUDA kernels for NVIDIA GPUs.
• Profile GPU workloads using advanced profiling tools.
• Collaborate with AI systems and model acceleration teams.

Skills

NVIDIA CUDA kernels optimization

GPU performance analysis

C++ proficiency

Memory management techniques

Tools

Nsight Compute

nvprof

CUDA-MEMCHECK

Job description

Pragmatike is hiring a CUDA Kernel Engineer to develop and optimize NVIDIA CUDA kernels for a leading AI startup. This remote role focuses on maximizing GPU performance and throughput for high-scale AI systems. Candidates should have substantial experience with CUDA and a profound understanding of GPU architecture. The position includes competitive salary, equity, sign-on bonuses, and comprehensive health benefits. Join a fast-paced environment that’s shaping the future of AI.