ML Infrastructure Engineer

Responsibilities: Build the infrastructure that lets Runsphere evaluate and secure AI behavior at high volume, with predictable latency and cost.

• Design and operate reliable pipelines for model inference, evaluation, and policy enforcement.
• Improve throughput, latency, observability, and cost across model-serving and data systems.
• Build tooling for model deployment, versioning, rollbacks, experiments, and reproducible runs.
• Support large-scale processing of prompts, tool calls, traces, and model outputs.
• Work with research and product engineers to move models and controls safely into production.
• Set strong standards for reliability, security, testing, and incident response.

Ideal candidate: You have built and operated production ML or data infrastructure and care about reliability, performance, and developer experience. You are comfortable with Python, cloud systems, containers, APIs, and distributed services, and can reason about model serving and evaluation workloads. You take ownership, communicate clearly, and enjoy hard systems problems. Experience with GPUs, inference optimization, orchestration, observability, or high-volume data pipelines is a plus.