GPU Software Engineer - AI 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

GPU Software Engineer - AI Acceleration Job SnapshotRole: GPU Software Engineer - AI AccelerationLocation: Abu Dhabi Emirate, United Arab EmiratesIndustry: Computer SoftwareFunction: IT-Software DevelopmentExperience: Strong experience in GPU programming, C , and performance optimizationJob Type: Contractor

Job DetailsCountry: United Arab EmiratesCity: Abu Dhabi EmirateIndustry: Computer SoftwareFunction: IT-Software DevelopmentSalary: 25000-45000Estimated salary range based on similar jobs in the job city; please confirm the final offer with the employer.Gender: AnyCandidate Nationality: AnyJob Type: Contractor

Role ContextThe GPU Software Engineer will work at the performance-critical layer between software workloads and modern graphics processing hardware. The assignment requires translating AI and LLM computational requirements into efficient GPU implementations while carefully managing kernel execution, host-device interaction, memory behavior, and hardware utilization. Strong low-level engineering judgment will help ensure workloads achieve measurable improvements in throughput, latency, and computational efficiency.

Key Responsibilities

• Design and implement GPU-accelerated software for AI, LLM, and compute-intensive workloads.
• Develop high-performance GPU functionality using CUDA, WebGPU, GLSL, or the most appropriate technology for each task.
• Profile GPU kernels and shaders to identify execution, memory, synchronization, and resource-utilization bottlenecks.
• Optimize kernels for improved throughput, latency, computational efficiency, and hardware utilization.
• Develop robust host-side C logic and integrate CPU-side applications with GPU workloads.
• Create technically rigorous GPU-focused tasks and reference solutions for AI and large language model applications.
• Analyze runtime behavior and use performance evidence to determine appropriate optimization strategies.
• Refactor performance-sensitive GPU and C code where architectural or implementation changes can improve efficiency.
• Evaluate memory access patterns, data movement, thread execution, and workload organization when diagnosing performance constraints.
• Optimize shader implementations used for graphics or general-purpose GPU computation.
• Benchmark revised implementations to confirm that optimization work produces measurable improvements.
• Apply GPU architecture knowledge when adapting workloads to NVIDIA hardware and other supported execution environments.
• Document implementation decisions, performance findings, optimization techniques, and technical outcomes clearly.

Ideal Profile

• Strong professional experience developing software for GPUs, particularly NVIDIA GPU platforms.
• Advanced C programming ability with experience building performance-sensitive applications.
• Practical proficiency in CUDA, WebGPU, GLSL, or a combination of these technologies.
• Detailed understanding of GPU architecture, parallel execution, memory hierarchy, and performance characteristics.
• Experience profiling and optimizing GPU kernels, compute workloads, or graphics shaders.
• Able to diagnose bottlenecks systematically and translate profiling findings into effective code-level improvements.
• Background in machine learning acceleration, graphics programming, scientific computing, high-performance computing, or another GPU-intensive discipline.
• Familiarity with host-device programming models and efficient CPU-GPU integration.
• Strong understanding of computational efficiency, memory optimization, parallelism, and workload scheduling.
• Comfortable tackling technically complex problems independently in a remote contractor environment.
• Able to communicate optimization strategies and engineering trade-offs precisely.

Skills Set

• GPU programming
• NVIDIA GPUs
• CUDA
• CUDA C
• C
• WebGPU
• GLSL
• GPU kernel development
• GPU kernel optimization
• Shader programming
• Compute shaders
• GPU performance profiling
• GPU architecture
• Parallel computing
• High-performance computing
• AI acceleration
• LLM acceleration
• Machine learning acceleration
• Scientific computing
• Graphics programming
• Host-device integration
• Memory optimization
• Performance benchmarking
• Bottleneck analysis
• GPU workload optimization

Why Join Us

This contractor role provides direct exposure to advanced GPU engineering for AI and large language model workloads, an area where specialized performance expertise continues to be highly valuable. The work combines low-level C , CUDA, shaders, parallel computing, and hardware-aware optimization rather than limiting the engineer to conventional application development. The remote structure also provides flexibility while maintaining a technically challenging focus on accelerated computing and measurable software performance.

About the Company

YO IT Consulting engages specialized technology professionals for software engineering, data, artificial intelligence, and advanced computing projects. The company connects domain experts with technically demanding assignments where practical programming knowledge, performance engineering, and high-quality implementation are essential to building and evaluating next-generation systems.