GPU Kernel Expert - AI Specialist

Mercor; San Francisco (CA); On-site; USD 150,000 - 210,000; Full time

Job summary

Mercor seeks a skilled evaluator to assess GPU/kernel tasks for frontier AI models. You will examine numerical correctness, performance benchmarks, and task scoping across diverse kernels, then deliver rubric-based feedback on methods and outcomes.

Responsibilities include verifying compilation/runtime validity and identifying failure modes, with emphasis on reproducibility and precision.

Qualifications

• 3+ years hands-on experience developing, optimizing, or verifying GPU/accelerator kernels in CUDA, Triton, NKI, or Pallas (JAX).
• Strong understanding of numerical-correctness criteria for kernels (absolute/relative/ULP tolerances).
• Experience with performance profiling and benchmarking (nsight, ncu, roofline analysis).
• Familiarity with common compilation and runtime failure modes.

Responsibilities

• Evaluate quality, correctness, and completeness of GPU/accelerator kernel development tasks used to train and evaluate a frontier AI lab's models.
• Assess numerical correctness, performance-benchmarking fairness, and task scoping across diverse kernel task types.
• Provide rubric-based written feedback on results and methods.

Skills

Numerical correctness

Performance profiling

Kernel development

Task scoping

Tools

CUDA

Triton

NKI

Pallas/JAX

Job description

Evaluate the quality, correctness, and completeness of GPU/accelerator kernel development tasks used to train and evaluate a frontier AI lab's models. You'll assess numerical correctness, performance-benchmarking fairness, task scoping, and compilation/runtime validity across diverse kernel task types — and provide clear, rubric-based written feedback.

Basic Qualifications

• 3+ years of hands-on experience developing, optimizing, or verifying GPU/accelerator kernels in at least two of: CUDA, Triton, NKI, or Pallas (JAX)
• Strong understanding of numerical-correctness criteria for kernels (absolute/relative/ULP tolerances, reference-implementation selection)
• Demonstrated experience with performance profiling and benchmarking (nsight, ncu, roofline analysis, or framework-native profilers)
• Familiarity with common compilation and runtime failure modes (driver mismatches, OOM, launch-configuration errors, shape/stride mismatches, autotuning failures)
• Experience with at least three kernel task types: generation from specification, translation/lowering across frameworks, migration between hardware targets, debugging, performance optimization, or operator fusion

Preferred Qualifications

• Experience across both NVIDIA GPU (CUDA/Triton) and custom-accelerator (NKI/Pallas/TPU) ecosystems
• Background in compiler engineering, MLIR, or intermediate-representation lowering
• Understanding of memory-hierarchy optimization (shared-memory tiling, register pressure, bank conflicts, coalescing patterns)
• Contributions to kernel libraries (cuBLAS, cuDNN, Triton community kernels, JAX/XLA custom calls)