Remote CUDA Kernel Engineer (Contract)

Congruex; Memphis (TN); On-site; USD 83,000 - 138,000; Full time

Job summary

micro1 is seeking a CUDA Engineering Expert contractor to optimize GPU kernels and contribute to a cutting-edge AI project. You will analyze, profile, and refactor CUDA/C++ and shader code, collaborating with remote teams to improve model training and inference performance.

This role emphasizes strong C++ and GPU programming skills, experience with GLSL and WebGPU, and proficiency with profiling tools like Nsight. Remote engagement, contract-based, with flexible hours and high-impact outcomes.

Qualifications

• Demonstrated CUDA programming expertise with kernel performance tuning.
• Advanced C++ development skills for high-performance computing.
• Hands-on GLSL and WebGPU shader development experience.
• Experience using GPU profilers for guided optimization.
• Strong analytical ability to evaluate kernel performance across hardware generations.
• Clear written and verbal communication for technical reporting.
• Experience collaborating in remote, cross-disciplinary project settings is a plus.

Responsibilities

• Analyze, profile, and optimize GPU kernels to maximize throughput on modern hardware.
• Collaborate with stakeholders to identify kernel bottlenecks and propose targeted optimizations.
• Refactor C++ and CUDA codebases for maintainability and efficiency across GPUs.
• Implement shader logic and graphics workflows using GLSL and WebGPU.
• Document findings, steps, and performance improvements with actionable reports.
• Contribute expertise to design discussions and evaluation of new GPU-based approaches.
• Stay informed on GPU programming advances and share relevant insights.

Skills

CUDA programming

C++ development

GLSL

WebGPU

GPU profiling

Tools

Nsight

Visual Profiler

Job description

micro1 is seeking a CUDA Engineering Expert contractor to optimize GPU kernels and contribute to a cutting-edge AI project. You will analyze, profile, and refactor CUDA/C++ and shader code, collaborating with remote teams to improve model training and inference performance.

This role emphasizes strong C++ and GPU programming skills, experience with GLSL and WebGPU, and proficiency with profiling tools like Nsight. Remote engagement, contract-based, with flexible hours and high-impact outcomes.