Software Engineer, Productivity - Inference Runtime

We're looking for an autonomous, high-ownership engineer who cares deeply about making other engineers faster, safer, and more confident. A major focus of this role will be improving the tooling and infrastructure around deploy gates for inference engine images. These systems help ensure that every image released to production and research is correct, numerically sound, free of regressions, and performant across key metrics like time-to-first-token (TTFT) and time-between-tokens (TBT). You'll help harden the systems that catch issues before they reach production, reduce noise from flaky or infrastructure-related test failures, and improve automation around triage, ownership, debugging, and escalation when failures occur. You'll also work on improving observability, rollout safety, release automation, and developer self-service tooling across a rapidly evolving inference stack. This is not generic internal tools work. The systems you build directly impact OpenAI's ability to support new model launches, safely ship inference optimizations to the world, onboard new infrastructure providers, and operate one of the largest and most performance-sensitive inference platforms in the world.

In this role, you will:

• Improve systems that ensure inference engine releases are correct, performant, and regression-free by evolving tooling and infrastructure for deploy gate validation

• Bring rigor to release, validation, branching, and deployment processes across the inference stack

• Improve canary, async, and large-scale validation workflows for inference systems

• Harden CI, testing, and validation infrastructure so failures are actionable and trustworthy

• Reduce noisy or flaky failures caused by infrastructure instability, GPU scheduling, or test environment issues

• Build automation for failure triage, ownership detection, debugging, and escalation

• Partner closely with inference teams, research developer productivity, engine acceleration, and infrastructure teams to improve release quality and rollout safety

• Reduce developer friction in testing, debugging, and release workflows so engineers can move faster with confidence

• You have strong experience with CI/CD systems, testing infrastructure, release tooling, developer productivity, or large-scale build and validation systems

• You are excited by high-impact infrastructure where small regressions in correctness, latency, or reliability meaningfully affect production systems

• You care about building systems engineers can trust, not just systems that technically function

• You have strong developer empathy and enjoy improving workflows, reducing friction, and making engineers more effective

• You demonstrate high ownership and proactively identify problems, drive improvements, and follow issues through resolution

• You are comfortable working in Python-heavy environments and debugging complex distributed systems

• You enjoy building automation that reduces manual triage, improves signal quality, and scales operational effectiveness

• You are comfortable operating in ambiguous areas without a fully predefined roadmap

• You enjoy partnering closely with engineers to understand workflows, pain points, and operational challenges

• You are pragmatic, collaborative, and motivated by helping teams move faster with more confidence

• You are excited to learn about large-scale inference systems, even if you have not worked directly on inference before

Python experience is highly relevant, as much of the current deploy gate and validation infrastructure is Python-based. C++ experience is helpful, especially for working near inference engine code, CI build issues, or performance-sensitive systems, but it is not required.

Prior inference experience is not required.

The ideal candidate is someone with strong instincts around developer productivity, testing, release engineering, and automation who is excited to apply those skills in a deeply impactful inference environment. We're looking for someone who is technically curious, comfortable navigating ambiguous, cross-functional operational problems, and is motivated to improve the reliability, safety, and developer experience of large-scale production infrastructure.