Member of the Technical Staff - Customer Tools

Role overview

Build the first version of software that lets AI labs operate compute clusters across multiple providers and locations as a single fleet. This individual contributor role focuses mainly on backend and infrastructure work, with a customer-facing feedback loop: you will help shape the product from real operational problems, ship features, join an on-call rotation, and speak with customers regularly.

Responsibilities

- Build services that connect clusters from different providers and customer sites under a unified way to manage and run jobs. - Create visibility into cluster health, GPU usage, spending, and alerts so customers can monitor their fleets in one place. - Develop organization, role, authorization, and single sign-on capabilities across customer clusters. - Help customers identify unused GPU capacity, offer it through a marketplace, and reclaim it when needed. - Work with senior engineers and customer research and infrastructure teams to define problems and deliver early product capabilities. - Participate in production on-call and improve the reliability of systems customers depend on.

Requirements

- Experience building multi-tenant systems with authorization, organizations, and separation of customer data. - Hands-on experience building against the Kubernetes API. - Experience creating telemetry pipelines for metrics, logs, or events that support dashboards and alerts. - Production backend experience in Go, or systems experience that would allow you to become productive in Go quickly. - Ability to build the frontend needed to deliver features you own; the product uses a TypeScript console. - Experience carrying an on-call pager for a production system and improving its operational reliability. - Comfort turning loosely defined customer problems into product decisions and shipped software.

Nice to have

Experience with machine-learning or high-performance computing infrastructure, training clusters, job schedulers, developer platforms, enterprise identity integrations such as SSO or SCIM, or moving large datasets across clusters or regions.

Benefits and work setup

The position is full time and open to North America remote candidates or candidates based in San Francisco. The listed package includes pay and equity, healthcare, dental and vision coverage, a retirement plan, and unlimited paid time off.