top of page
< Back

GPU Kernel Expert

$70–$90/h

United States

Role overview

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)

Similar jobs

AI Developer Trace Task Auditor · Mercor · $70–$90/h · United States

AWS Serverless & Infrastructure-as-Code Task Auditor · Mercor · $70–$90/h · United States

Application Users - Android Studio on macOS - Code · Mercor · $55–$65/h · Remote

You’ll be redirected to our partner’s application page

AI TRAINING OPPORTUNITY

bottom of page
Sporterer
OPEN OPPORTUNITIES

AI Training Opportunities

Loading opportunities…

Sporterer
Future of work · September 2026

How big is the remote / hybrid opportunity?

Remote work is already huge — but the bigger story may be the hundreds of millions of jobs that could still move beyond geography.

Global remote and hybrid work opportunity infographic

Remote work already connects millions of people with companies they would never have been able to work for a decade ago. But compared with the size of the global labour market, it may still be early.

The interesting number is not only how many people work remotely today — it is how many roles could plausibly become remote or hybrid next.

The chart above starts with roughly 3.3 billion people employed worldwide. Around 300 million jobs are already remote or hybrid, while an additional pool of roughly 250 million roles could potentially move in that direction as technology, workflows and company culture continue to change.

Skills are becoming more portable

This matters because more work can increasingly be matched to expertise rather than location. AI training and evaluation are a good example: companies need lawyers, engineers, designers, editors, researchers, finance professionals and many other specialists to help improve AI systems.

For workers, that creates access to projects outside their local market. For companies, it expands the talent pool from one city or country to the world.

Explore remote opportunities →