Hirely is recruiting on behalf of a hiring partner for a remote CUDA / GPU Engineering chance supporting a project with a leading AI research organization. Contributors will optimize and evaluate GPU kernels across modern hardware environments, using performance profiling and low-level GPU programming expertise to improve efficiency, utilization, and execution speed.

What You’ll Do:

- Analyze and optimize GPU kernels for performance, efficiency, and hardware utilization.
- Use GPU profiler metrics such as L2 cache hit rate, L2 throughput, occupancy, and related performance signals to identify bottlenecks.
- Review existing GPU kernel implementations and determine optimization opportunities.
- Write, modify, and evaluate C++17, Python, and GPU programming code .
- Apply CUDA, HIP, shader programming, or related GPU programming models to improve kernel performance.
- Evaluate kernel behavior across modern GPU architectures.
- Document optimization decisions and explain performance tradeoffs clearly.
- Work on challenging performance-engineering problems where strong GPU expertise is required.

Who We’re Looking For:

- 1+ year of professional or graduate-level GPU programming/research experience.
- Strong proficiency in C++ through C++17 .
- Working knowledge of Python and Git .
- Experience with at least one GPU programming model, including: CUDA, HIP, Slang, HLSL, GLSL or similar GPU/kernel programming technologies
- Strong understanding of GPU performance profiling and optimization .




- Experience analyzing GPU performance metrics and identifying kernel bottlenecks.
- Ability to optimize GPU kernels without requiring extensive knowledge of the underlying application.
- Availability for 20+ hours per week .

Preferred Qualifications:

- Experience with CUDA C++ Core Libraries.
- Inline PTX or low-level GPU programming experience.
- Tensor Core optimization experience.
- NVIDIA Blackwell optimization experience.
- Experience with NVIDIA Nsight Compute .
- Background at NVIDIA, AMD, Qualcomm, or another GPU hardware/software organization.
- Open-source contributions involving GPU programming or kernel optimization.

Compensation & Time Commitment:

- $500 per accepted task
- 20+ hours per week
- Fully remote

Application Process: Qualified applicants may go through resume screening and a technical assessment or additional technical evaluation before onboarding. Candidates should be prepared to demonstrate practical GPU programming and kernel optimization expertise.

About Hirely:

Hirely connects qualified professionals with remote opportunities offered by hiring partners across Artificial Intelligence, Software Engineering, GPU Computing, Machine Learning, Research, and other specialized technology fields. Hirely may serve as a recruitment and talent-sourcing partner. Final hiring decisions, onboarding, compensation, and project assignments are determined solely by the hiring partner.

📌 CUDA Engineering Expert | $500/Task (India)
🏢 Hirely
📍 India