CUDA Engineering Expert

Mercor is seeking GPU kernel optimization experts to contribute to a project with a leading AI lab.

This opportunity is designed for professionals with strong C++ skills, practical GPU programming experience, and expertise in optimizing kernels using profiler-guided analysis.

Contributors will evaluate, optimize, and reason about GPU kernels across modern hardware environments, helping improve performance and efficiency on advanced GPU architectures.

This role is ideal for engineers who enjoy low-level performance optimization and maximizing hardware utilization.

Key Responsibilities

• Analyze and optimize GPU kernels for:

  • Performance
  • Efficiency
  • Hardware utilization

• Use profiler metrics such as:

  • L2 cache hit rate
  • L2 throughput
  • Occupancy
  • Memory performance indicators
  • Compute utilization metrics

• Review GPU kernel implementations and identify performance bottlenecks

• Write, modify, and reason about:

  • C++17 code
  • Python code
  • GPU programming code

• Apply expertise in:

  • CUDA
  • HIP
  • Shader programming
  • Related GPU programming models

• Document optimization decisions and explain profiler-guided performance improvements

• Evaluate when specific profiler metrics are useful and when they are not

Required Qualifications

• Availability of at least 20 hours per week

• Strong C++ skills through C++17

• Working knowledge of:

  • Python
  • Git

• Fluency in at least one GPU programming model:

  • CUDA
  • HIP
  • Slang
  • HLSL
  • GLSL
  • Similar GPU programming frameworks

• At least one year of:

  • Professional GPU experience
  • Graduate-level GPU research experience

• Strong understanding of:

  • GPU architecture
  • Kernel optimization
  • GPU profiler metrics
  • Performance analysis workflows

• Ability to optimize kernels without requiring deep context on every underlying algorithm

Preferred Qualifications

• Experience with:

  • CUDA
  • HIP
  • CUDA C++ Core Libraries
  • Inline PTX assembly
  • Tensor Core optimization

• Experience optimizing for:

  • NVIDIA Blackwell GPUs

• Familiarity with:

  • NVIDIA NSight Compute

• Prior experience with GPU-focused organizations such as:

  • NVIDIA
  • AMD
  • Qualcomm

• Open-source contributions involving GPU optimization

Application Process

• Submit a resume or relevant technical background
• Qualified candidates may be asked to:
  • Complete a technical assessment
  • Provide additional technical information

Contract & Payment Terms

• Independent contractor engagement
• Fully remote work
• Flexible schedule
• Weekly payments through Stripe or Wise
• Projects may be extended, shortened, or concluded based on project needs and performance

Important Note

Mercor currently cannot support:

• H1-B candidates
• STEM OPT candidates

About Mercor

Mercor partners with leading AI labs and enterprises to train and improve frontier AI systems using human expertise.

Contributors work directly on projects helping shape the next generation of AI-powered technologies.