MLOps Engineer: JAX/PyTorch & GPU Kernel Expert — Remote

Weekday AI (YC W21); United States; On-site; USD 96,000 - 152,000; Full time

Job summary

Weekday AI (YC W21) seeks an experienced MLOps Engineer for a remote, full‑time engagement aimed at advancing Generative AI infrastructure. You’ll collaborate with researchers and engineers to design, implement, and optimize large‑scale training systems, ML pipelines, and kernel‑level performance using JAX, PyTorch, Pallas, and Triton.

Requirements include 2+ years in MLOps or ML infra, strong written communication, and a proven ability to deliver production‑grade ML infrastructure.

Qualifications

• Minimum 2 years of professional experience in MLOps or ML systems.
• Hands-on production experience with JAX and/or PyTorch.
• Experience developing or optimizing custom GPU kernels using Pallas (JAX) or Triton.
• Strong understanding of distributed training systems and scalable ML infrastructure.
• Excellent written communication skills.
• Availability to work 40 hours per week during standard weekday business hours.

Responsibilities

• Collaborate with research and engineering teams on MLOps, ML infrastructure, and large-scale training systems.
• Design challenging real-world ML systems tasks reflecting production scenarios.
• Develop accurate, well-documented solutions to ML infra and training pipeline problems.
• Review and evaluate technical tasks and AI-generated solutions with clear feedback.
• Create evaluation rubrics for distributed training, ML pipelines, and kernel-level work.
• Collaborate with SMEs to maintain consistency and quality across training datasets.
• Contribute domain expertise to improve AI reasoning.

Skills

MLOps

JAX

PyTorch

Distributed training

Kernel programming

ML infrastructure

Strong communication

Tools

Pallas

Triton

Job description

Weekday AI (YC W21) seeks an experienced MLOps Engineer for a remote, full‑time engagement aimed at advancing Generative AI infrastructure. You’ll collaborate with researchers and engineers to design, implement, and optimize large‑scale training systems, ML pipelines, and kernel‑level performance using JAX, PyTorch, Pallas, and Triton.

Requirements include 2+ years in MLOps or ML infra, strong written communication, and a proven ability to deliver production‑grade ML infrastructure.