Software Engineer, API Multicloud

About the Team

API Multicloud partners with cloud service providers (CSPs) to bring OpenAI’s frontier models and APIs into the cloud environments where developers and enterprises already build. We own the API end to end across CSP platforms, from API integration and request routing through inference execution, capacity management, and day-to-day operations. Our mission is to make these capabilities available broadly and safely while preserving the reliability and developer experience customers expect from OpenAI.

Delivering that experience requires work across the stack: adapting API services and cloud-native integrations, bringing up model-serving environments, validating new releases, and operating the systems that connect customer requests to available capacity. We work directly with CSP engineering teams and partner closely with OpenAI’s inference, API, infrastructure, security, and Safety Systems teams. Together, we turn new model capabilities and partner requirements into dependable production services, with attention to performance, operational readiness, and efficient use of compute.

Our work contributes to OpenAI’s AWS partnership , including the launch of Daybreak Red and Daybreak Blue on Amazon Bedrock . We’re building on that foundation to bring OpenAI’s API to additional CSPs. These efforts let customers use OpenAI capabilities within the cloud infrastructure, security controls, and workflows they already rely on.

About the Role

We’re looking for experienced infrastructure engineers to design, build, and operate the systems that run OpenAI’s API across CSP platforms. You’ll help establish the cloud deployment platform: configuring and launching model-serving environments, planning capacity, improving caching and utilization, and making new releases safe and repeatable. Working closely with inference engineers, you’ll connect advances in model serving to the infrastructure and operational practices needed to run them reliably in production.

The role calls for strong distributed systems judgment and comfort moving across API services, cloud infrastructure, deployment tooling, and production debugging. You’ll work directly with CSP engineers to navigate differences in infrastructure, native services, and operating models, resolve integration issues, and build reusable solutions that reduce repeated work as we expand. Your decisions will affect how quickly new capabilities reach customers, how reliably they run, and how efficiently we use cloud resources.

In this role, you will:

Design and build cloud infrastructure and distributed services that connect API requests, routing, and model-serving deployments.

Stand up and operate model-serving environments in partnership with inference engineers, bringing new models and serving capabilities into production.

Plan and manage cloud capacity, improving utilization, caching, request performance, and cost efficiency as demand grows.

Create repeatable build, release, and validation processes that catch functional and performance regressions before launch.

Develop reusable infrastructure abstractions that let API capabilities run consistently across CSPs while accommodating provider-specific requirements.

Partner with OpenAI’s inference, API, infrastructure, and security teams and CSP engineers to bring up the API serving stack on new platforms, resolve operational issues, and shape shared technical priorities.

Own production health through observability, incident diagnosis, reliability improvements, and practical automation.

Your background might look something like:

7+ years of experience leading cloud infrastructure or distributed systems projects from architecture through launch and ongoing production ownership.

Strong distributed systems fundamentals and the ability to reason about performance, reliability, and failure modes.

Hands-on experience building and operating services on a major CSP, including deployment systems, networking, storage, and cloud-native integrations.

Strong production coding and debugging skills, with the ability to move confidently through unfamiliar services and infrastructure.

Experience with capacity planning, performance optimization, or operational tooling for production systems at scale.

Sound technical judgment: you can turn ambiguous requirements into practical systems and reusable infrastructure.

Clear technical communication and experience working across teams or organizational boundaries to deliver complex projects.

Curiosity about model serving and AI infrastructure. Experience deploying models is a plus; direct ML or model-training experience is not required.