GPU Kernel Architect & Quality Validator

Dorado; Northern (KY); Hybrid; USD 100,000 - 180,000; Full time

Job summary

Dorado is seeking a scientist-engineer to evaluate GPU/accelerator kernel development tasks used to train and evaluate frontier AI models. You will assess numerical correctness, performance benchmarks, task scoping, and runtime validity across diverse kernel types.

You will deliver rubric-based written feedback, identify failure modes, and help shape best practices for future kernel work in a collaborative research environment.

Qualifications

• 3+ years hands-on kernel development or verification.
• Experience with CUDA, Triton, NKI, or Pallas (JAX).
• Familiarity with runtime failures (OOM, driver issues, shape/stride).

Responsibilities

• Evaluate kernel tasks for correctness, performance, and completeness.
• Provide rubric-based feedback on scope, accuracy, and quality.
• Identify and document compilation/runtime failures across frameworks.
• Recommend improvements and best practices for kernel development.

Skills

Numerical correctness

Performance profiling

Kernel development

Tools

CUDA

Triton

NKI

Pallas (JAX)

Job description

Dorado is seeking a scientist-engineer to evaluate GPU/accelerator kernel development tasks used to train and evaluate frontier AI models. You will assess numerical correctness, performance benchmarks, task scoping, and runtime validity across diverse kernel types.

You will deliver rubric-based written feedback, identify failure modes, and help shape best practices for future kernel work in a collaborative research environment.