MLOps Engineer (JAX, PyTorch, Pallas/Triton)

Role Overview

MLOps Engineer to support a cutting-edge Generative AI / LLM initiative. You will create, evaluate, and refine technical tasks that improve large-scale training and model reasoning across MLOps and ML infrastructure.

Responsibilities

• Partner with research and engineering teams to strengthen capabilities across MLOps, ML infrastructure, and large-scale training systems.
• Design challenging, production-like MLOps and ML systems tasks.
• Develop accurate, well-documented solutions for complex training pipeline and infrastructure problems.
• Review and evaluate technical work (including AI-generated solutions) and provide clear written feedback.
• Create evaluation rubrics and scoring frameworks covering:
  • Distributed training architectures
  • ML pipeline design
  • Infrastructure optimization
  • Kernel-level programming
  • Performance tuning
• Collaborate with subject matter experts to maintain consistency and technical accuracy.
• Contribute domain expertise to improve advanced AI system reasoning.

Requirements

• 2+ years of professional experience in MLOps, ML Infrastructure, or ML Systems Engineering.
• Hands-on production experience with JAX and/or PyTorch in large-scale ML environments.
• Practical experience developing/optimizing custom GPU kernels using Pallas (JAX) and/or Triton.
• Strong understanding of distributed training, model optimization, and scalable ML infrastructure.
• Ability to work 40 hours/week during standard weekday business hours.
• Excellent written communication skills.

Preferred Qualifications

• Experience designing/optimizing large-scale ML training pipelines.
• Knowledge of distributed computing and GPU performance optimization.
• Familiarity with AI/ML evaluation methodologies.
• Experience collaborating with research teams on advanced ML projects.

Contract & Logistics

• Independent contractor engagement (independent contractor basis).
• Fully remote with full weekday availability (40 hours/week).