Trainium NKI Kernel Specialist

Mercor; San Francisco (CA); Remote; USD 120,000 - 150,000; Full time

Job summary

Mercor seeks an experienced evaluator to scrutinize Neuron Kernel Interface (NKI) development tasks used to train and evaluate frontier AI models. You will assess CUDA→NKI migration fidelity, Trainium-specific optimization quality, and cross-platform numerical correctness, delivering rubric-based written feedback for engineering teams.

The role emphasizes kernel-level analysis, memory-hierarchy management, and performance bottlenecks across Trainium hardware while ensuring standards are met for

Qualifications

• 2+ years hands-on experience developing or optimizing kernels using NKI targeting AWS Trainium/Inferentia2 hardware.

Responsibilities

• Assess quality, correctness, and hardware-appropriateness of NKI development tasks.
• Provide rubric-based written feedback on migration fidelity and performance.
• Evaluate cross-platform numerical correctness (GPU vs Trainium) and reporting.

Skills

NKI kernel development

CUDA→NKI migration evaluation

Trainium performance profiling

GPU numerical-correctness standards

DMA orchestration & memory hierarchy

Tools

AWS Neuron SDK

Neuron Compiler

CUDA

Triton kernel development

Job description

Mercor seeks an experienced evaluator to scrutinize Neuron Kernel Interface (NKI) development tasks used to train and evaluate frontier AI models. You will assess CUDA→NKI migration fidelity, Trainium-specific optimization quality, and cross-platform numerical correctness, delivering rubric-based written feedback for engineering teams.

The role emphasizes kernel-level analysis, memory-hierarchy management, and performance bottlenecks across Trainium hardware while ensuring standards are met for