Remote GPU Software Engineer - AI/LLM Acceleration

YO IT Consulting; Al Ruways Industrial City; On-site; AED 279,000 - 502,000; Part time

Benefits offered by this job

Remote contractor

Flexible hours

Job summary

YO IT Consulting is seeking a GPU Software Engineer - AI Acceleration in Abu Dhabi to design and implement GPU-accelerated software for AI workloads. You will work at the performance-critical layer between software workloads and modern graphics processing hardware, translating AI and LLM requirements into efficient GPU implementations.

The role emphasizes low-level C, CUDA, shaders, and performance optimization, with remote contractor arrangement and opportunities to influence throughput and

Qualifications

• Strong professional experience developing software for GPUs, particularly NVIDIA GPU platforms.
• Advanced C programming ability with experience building performance-sensitive applications.
• Proficiency in CUDA, WebGPU, GLSL, or a combination of these technologies.
• Understanding of GPU architecture, parallel execution, memory hierarchy, and performance characteristics.
• Experience profiling and optimizing GPU kernels, compute workloads, or graphics shaders.

Responsibilities

• Design and implement GPU-accelerated software for AI, LLM, and compute-intensive workloads.
• Develop high-performance GPU functionality using CUDA, WebGPU, GLSL, or equivalent.
• Profile GPU kernels and shaders to identify bottlenecks and optimize performance.
• Optimize kernels for throughput, latency, and hardware utilization.
• Develop robust host-side C logic and integrate CPU-side applications with GPU workloads.
• Document implementation decisions and performance outcomes clearly.

Skills

GPU programming

CUDA

WebGPU

GLSL

C

NVIDIA GPUs

Shader programming

Performance profiling

Memory optimization

Parallel computing

LLM acceleration

AI acceleration

Tools

CUDA toolkit

Job description

YO IT Consulting is seeking a GPU Software Engineer - AI Acceleration in Abu Dhabi to design and implement GPU-accelerated software for AI workloads. You will work at the performance-critical layer between software workloads and modern graphics processing hardware, translating AI and LLM requirements into efficient GPU implementations.

The role emphasizes low-level C, CUDA, shaders, and performance optimization, with remote contractor arrangement and opportunities to influence throughput and