Backend Software Engineer, GTM Innovation

OpenAI; United States; Remote; USD 140,000 - 210,000; Full time

Job summary

OpenAI invites a product-minded backend engineer to join the GTM Innovation team in the United States to build the systems behind a new generation of AI coworkers—persistent agents that understand their environment and take on meaningful work.

You will design and ship foundational infrastructure for durable workflow orchestration, memory and context systems, and the tools, permissions, and evaluation frameworks needed to operate reliably, with strong observability and continuous improvement.

Responsibilities

• Build and scale durable agent runtimes and long-running agent loops.

Job description

About the Team: GTM Innovation’s mission is to automate 100% of digital knowledge work for OpenAI's GTM so sellers spend more time directly with customers applying their strategy and judgment. AGI-level reasoning doesn’t mean organization-level transformation “just works” out of the box; orgs must be redesigned around abundant intelligence and persistent virtual coworkers. The team builds and scales a fleet of virtual coworkers that operate as full-time employees and move increasingly ambitious tasks from the foreground to the background.

About the Role We’re looking for a product‑minded backend engineer to build the systems behind a new generation of AI coworkers: persistent agents that understand their environment, take on meaningful work, and improve over time. You’ll design and ship the foundational infrastructure that makes these agents effective in production, from durable workflow orchestration to long‑running agent loops built on the Codex harness to the context, memory, tools, and permissions model they need to act reliably. You’ll also build rigorous evaluation, observability, and feedback systems that make agent quality measurable and continuously improve how these systems perform. This is a hands‑on, high‑ownership role at the intersection of agent infrastructure, applied AI, and product engineering. You’ll work directly with go‑to‑market teams and collaborate across OpenAI’s product and research organizations to turn ambiguous, real‑world needs into dependable systems that help OpenAI meet the world at scale.

In this role, you will:

• Build and scale durable agent runtimes, long-running agent loops, and workflow orchestration using the Codex harness and emerging agent infrastructure.
• Design context and memory systems that ground agents in the information they need while preserving clear access boundaries.
• Develop rigorous evaluation frameworks, production feedback loops, and observability to measure agent quality and improve it over time.
• Partner directly