Remote: Trainium NKI Kernel Expert for AI Model Tuning

Dorado; Northern (KY); Hybrid; USD 120,000 - 180,000; Full time

Benefits offered by this job

Job summary

Dorado seeks a specialized evaluator to review Neuron Kernel Interface (NKI) development tasks used to train and evaluate frontier AI models. You will judge CUDA→NKI migration fidelity, Trainium-specific optimization quality, and cross-platform numerical-correctness standards, delivering rubric-based feedback.

Responsibilities include thorough analysis of kernel patterns, memory-hierarchy use, and DMA orchestration, with clear recommendations for improvements and validation criteria aligned to

Qualifications

• 2+ years hands-on kernel development or optimization with NKI on Trainium/Inferentia2
• Experience with NKI patterns: tile-based compute, DMA, memory hierarchy
• Experience evaluating CUDA→NKI migrations and accuracy

Responsibilities

• Evaluate NKI development tasks for quality and hardware-appropriateness
• Provide rubric-based feedback on performance and correctness
• Assess CUDA→NKI migrations and Trainium-specific optimization quality

Skills

NKI kernel development

CUDA→NKI migration

Trainium hardware

Memory hierarchy

Profiling & optimization

Numerical correctness

Tools

AWS Neuron SDK

Neuron Compiler

Job description

Dorado seeks a specialized evaluator to review Neuron Kernel Interface (NKI) development tasks used to train and evaluate frontier AI models. You will judge CUDA→NKI migration fidelity, Trainium-specific optimization quality, and cross-platform numerical-correctness standards, delivering rubric-based feedback.

Responsibilities include thorough analysis of kernel patterns, memory-hierarchy use, and DMA orchestration, with clear recommendations for improvements and validation criteria aligned to