Software Engineer, Agent Platform

OpenAI•San Francisco, CA

3d•$230,000 - $385,000•Hybrid

About The Position

The Agent Platform team builds production systems that enable AI agents to execute code, use tools, and interact with software reliably and securely at scale. The team works with product, research, and infrastructure teams to turn new agent capabilities into dependable product experiences, including execution environments, APIs, and integrations supporting products such as Codex and tool use in ChatGPT. As a Software Engineer on the team, you will build and scale services for OpenAI’s agentic products, bringing capabilities from prototypes into production. The work includes designing APIs and services, improving agent execution and orchestration, and building integrations. You will help launch new experiences while improving platform reliability, observability, and efficiency. The role is based in San Francisco, California, with a hybrid schedule of three days in the office per week. Relocation assistance is offered to new employees.

Requirements

Experience building and operating large-scale backend services, distributed systems, or cloud infrastructure.
Strong understanding of system performance, concurrency, failure modes, and resource management.
Comfort working across service boundaries and collaborating with teams to solve problems end to end.

Nice To Haves

Enjoy building shared platforms and APIs that help other engineers move faster.
Ability to turn an early prototype into a reliable service and evolve it as usage grows.
Care about how infrastructure decisions affect user experience, including responsiveness, reliability, and predictable behavior.
Experience with cloud platforms, containers, orchestration systems, or infrastructure-as-code tools such as Terraform.
Interest in secure execution environments, sandboxing, and the challenges of running agent workloads in production.
Comfort moving quickly through ambiguous problems while making thoughtful decisions about security, performance, and long-term maintainability.

Responsibilities

Build and operate production services that enable agents to execute code, use tools, and interact with external systems.
Design APIs and platform abstractions that help product teams build and launch agentic experiences.
Develop orchestration systems for agent execution, state, resource allocation, retries, and recovery.
Improve platform reliability, latency, and efficiency through performance optimization, observability, and capacity planning.
Partner with product, research, security, and infrastructure teams to bring new agent capabilities into production.
Own systems throughout their lifecycle, from design and implementation through rollout, monitoring, and ongoing operation.