GenAI MLOps GPU Engineer - High-Throughput Inference

Obsidian; San Francisco (CA); On-site; USD 120,000 - 180,000; Full time

Job summary

Cincinnatus LLC is seeking MLOps Engineers to join a cutting-edge GenAI team building foundational AI models. This hands-on role focuses on large language model infrastructure, GPU kernels, performance profiling, and high-throughput inference serving, with opportunities to contribute to training data for frontier AI systems.

This is a 40-hour per week, full-time engagement as the employer-of-record, with placement at a leading AI Lab.

Qualifications

• 2+ years hands-on experience in ML systems, ML infrastructure, or model serving.
• Hands-on in GPU kernels, profiling tools, distributed workloads, or high-throughput inference.
• Production experience with JAX and/or PyTorch is required.

Responsibilities

• Design tasks across GPU kernels, profiling, debugging, and inference serving; provide detailed solutions.
• Guide teams to improve model performance on training infrastructure and framework topics.
• Evaluate MLOps solutions and provide clear, written feedback.
• Develop evaluation rubrics for kernel optimization, profiler outputs, and serving latency.
• Collaborate with experts to ensure training data quality and consistency.

Skills

ML systems

GPU kernel programming

Performance profiling

Distributed workloads

PyTorch/JAX

Technical communication

Tools

Kineto

Torch Profiler

Nsight

XLA

CUDA

Job description

Cincinnatus LLC is seeking MLOps Engineers to join a cutting-edge GenAI team building foundational AI models. This hands-on role focuses on large language model infrastructure, GPU kernels, performance profiling, and high-throughput inference serving, with opportunities to contribute to training data for frontier AI systems.

This is a 40-hour per week, full-time engagement as the employer-of-record, with placement at a leading AI Lab.