Hirely
Website:
onhirely.com
Company:
https://www.linkedin.com/company/hirelyglobal
Seniority: Entry level
Industries: Staffing and Recruiting
Job details:
Hirely is recruiting for remote CUDA Engineering opportunities with our hiring partners.
These projects support the development of next-generation Artificial Intelligence (AI), Machine Learning (ML), High-Performance Computing (HPC), and GPU-accelerated computing platforms. Depending on the hiring partner and project requirements, you may contribute to CUDA kernel optimisation, GPU programming, performance engineering, AI infrastructure, scientific computing, or large-scale parallel computing initiatives. Compensation, contract terms, schedules, and project scope vary depending on the hiring partner and assignment.
Responsibilities:
- Design, develop, optimise, and maintain high-performance GPU applications using CUDA C++ and related technologies.
- Analyse GPU kernels and identify performance bottlenecks using profiling tools and hardware performance metrics.
- Optimise memory access, kernel execution, occupancy, throughput, cache utilisation, and overall GPU efficiency.
- Develop, modify, and review C++, CUDA, Python, or GPU-related code while following software engineering best practices.
- Apply GPU programming techniques using CUDA, HIP, OpenCL, or similar parallel computing frameworks where applicable.
- Collaborate with engineering teams to improve AI workloads, scientific computing applications, and high-performance software systems.
- Document optimisation approaches, technical decisions, and performance improvements clearly and accurately.
- Participate in debugging, testing, code reviews, and continuous performance improvements.
Preferred Qualifications:
- Professional experience in CUDA development, GPU programming, High-Performance Computing (HPC), Parallel Computing, AI Infrastructure, Graphics Programming, or related engineering disciplines.
- Strong proficiency in modern C++ (C++17 or later) and practical experience with Python and Git.
- Experience developing GPU kernels using CUDA or related GPU programming technologies such as HIP, OpenCL, HLSL, GLSL, or Slang.
- Strong understanding of GPU architecture, memory hierarchy, kernel optimisation, occupancy, cache behaviour, throughput, and parallel computing principles.
- Experience using GPU profiling and performance analysis tools such as NVIDIA Nsight Compute or equivalent.
- Familiarity with NVIDIA, AMD, Qualcomm, or similar GPU hardware platforms is beneficial.
- Open-source contributions, research experience, or advanced academic work related to GPU computing, CUDA optimisation, AI systems, or scientific computing is advantageous.
- Excellent analytical, debugging, problem-solving, and written communication skills.
Compensation:
Compensation normally ranges from $80 to $100 per hour. Project duration, workload, and contract terms are determined by the hiring partner.
Application Process:
Depending on the hiring partner's process, applicants may complete resume screening, AI-powered interviews, technical assessments, or other role-specific evaluations before a final hiring decision.
About Hirely:
Hirely is a recruitment and talent sourcing platform connecting professionals with remote opportunities from trusted hiring partners across AI, software engineering, data, research, healthcare, finance, engineering, and other professional industries.
Important Notice:
Hirely may act as a recruitment or talent sourcing partner for this opportunity and may not be the direct employer. Employment terms, compensation, onboarding, and final hiring decisions are determined by the hiring partner responsible for the position.
Click on Apply to know more.