Machine Learning Infrastructure Engineer

What you'll do

• Manage compute infrastructure for research and production model training and serving, optimizing latency, throughput, accelerator costs, and memory use.
• Support inference for evaluation, synthetic data, and reinforcement learning; build model release and observability workflows.
• Coordinate GPU capacity and model rollouts in customer environments with founders and researchers.

What they ask for

• Professional experience at a company building or tuning production GPU infrastructure for large language models; coursework or personal projects alone do not qualify.
• Deep expertise in diagnosing GPU performance limits, tuning model serving, and setting up monitoring.
• Hands-on experience with an LLM serving framework or comparable in-house engine, using PyTorch or JAX.
• Solid debugging and low-level systems skills, understanding accelerator memory and compute limits, and a track record of owning infrastructure systems end to end.

Nice to have

• Experience at an AI-focused startup that trains its own models.
• Low-level optimization experience, a technical degree from a leading program, or publications at top machine learning conferences or public writing about systems work.

Pay

USD 200,000 to 350,000 per year (base salary)

Tools

GPU, PyTorch, JAX

AI focus

Infrastructure for training and serving language models, including inference for evaluation, synthetic data creation, and reinforcement learning workloads.