ML Infrastructure Engineer

We are looking for an ML Infrastructure Engineer with 3+ years of experience to own and scale the training and inference stack at a fast-growing AI document processing platform. You'll be a strong generalist who understands the mechanics of how ML models work – from serving and monitoring to building robust data pipelines – and can improve inference performance, reliability, and cost efficiency. This is a high-impact IC role where you'll work closely with ML researchers to ensure models are deployed quickly and reliably, and that infrastructure is never a bottleneck for the products being served. The ideal candidate is AI-native from the get-go, comfortable with 1-to-3 node training and single-to-double node serving, and thrives in a fast-paced startup environment.

What you will be doing

- Building and maintaining model serving infrastructure – improving inference speed, monitoring, and reliability to ensure it's never a bottleneck for customers

- Setting up and improving training infrastructure for models ranging from 300M to 30B parameters across 1-to-3 node environments

- Developing observability, logging, and monitoring systems across the ML stack

- Building internal data pipelines and tooling to help ML researchers move faster from experiment to production

- Architecting infrastructure to arbitrate inference between multiple cloud providers while optimizing for accuracy, latency, and cost