Other open roles

View all opportunities

GPU Kernel Expert

$90
per hour
Contractor Remote Last verified 21 Jul 2026
Affiliate disclosure

Work Expert (WE) is an independent publishing and referral website. We are not a recruiter, hiring manager, agent or employer, and we are not affiliated with or endorsed by Mercor. Applying takes you to the platform's own website, where we may be recorded as the referring source. We may receive a referral fee at no additional cost to you. Read our full affiliate disclosure →

About the Role

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.

What You'll Do

    You're a Good Fit If You

    • 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
    • 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)

    Role Highlights

    Pay
    $70-$90/h
    Contract
    Contractor
    Location
    Remote
    Category
    Model evaluation
    Posted
    2 months ago
    Job ID
    WE-7849

    Pay & Payout

    Independent contractor engagement. Apply via the link below - Mercor handles onboarding and pays weekly via Stripe or Wise.

    Common Questions

    Mercor has a reliability rating of very reliable based on the listings we track, with an onboarding time of 1-3 weeks (interview + assessment + trial). See our full Mercor review for the details behind that rating.

    Applying takes you to Mercor's own website, where the application is completed. We are not a recruiter or employer and do not process applications directly - we may be recorded as the referring source.

    $70-$90/h - Independent contractor engagement. Apply via the link below - Mercor handles onboarding and pays weekly via Stripe or Wise.

    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.

    Apply on Mercor

    Opens Mercor in a new tab. You pay nothing; we may earn a referral fee.

    Get similar jobs every week

    What are you interested in?
    Model evaluation

    We only use your email to send job alerts. Read our Privacy Policy.