Senior ML Performance Engineer - GPU Inference & CUDA

Mixpeek; New York (NY); On-site; USD 140,000 - 220,000; Full time

Job summary

Modal is seeking strong engineers to build and optimize ML systems at scale. You will contribute to the container runtime and open-source projects, pushing language models and diffusion models toward higher throughput and lower latency on GPU-enabled infrastructure.

The role emphasizes performance engineering, profiling, and system-level improvements. Ideal candidates have 5+ years of high-quality, high-performance code experience, deep knowledge of PyTorch/ML frameworks, CUDA, and GPU

Qualifications

• 5+ years of experience writing high-quality, high-performance code.
• Experience with torch, ML frameworks, and inference engines (vLLM or TensorRT).
• Familiarity with Nvidia GPU architecture and CUDA.
• Experience with ML performance engineering and debugging GPU performance issues.
• Nice-to-have: familiarity with low-level OS foundations (Linux kernel, containers).

Skills

High-performance coding

Torch & ML frameworks

CUDA & GPU programming

ML performance engineering

Linux kernel familiarity

Job description

Modal is seeking strong engineers to build and optimize ML systems at scale. You will contribute to the container runtime and open-source projects, pushing language models and diffusion models toward higher throughput and lower latency on GPU-enabled infrastructure.

The role emphasizes performance engineering, profiling, and system-level improvements. Ideal candidates have 5+ years of high-quality, high-performance code experience, deep knowledge of PyTorch/ML frameworks, CUDA, and GPU