Software Engineer, CI/CD & ML Infrastructure

The role

You will define and build CI/CD systems that qualify frontier models and the software stack for custom AI accelerators. The role involves establishing scalable GitOps and monorepo practices and designing regression and benchmarking pipelines.

What you will do

• Define the model CI strategy for frontier models running on OpenAI custom silicon and integrated accelerator systems.
• Build production CI/CD pipelines for functional regression, performance benchmarking, software stability, and release qualification.
• Design representative test matrices across models, configurations, hardware generations, software components, and deployment environments.
• Create reliable performance baselines, regression detection, bisect and triage workflows, and clear ownership for failures.
• Establish GitOps practices for reproducible configuration, promotion, rollback, auditability, and environment consistency.
• Shape monorepo architecture, dependency management, build and test boundaries, change validation, and developer workflows at scale.
• Develop scalable orchestration, artifact management, caching, scheduling, observability, and capacity controls for accelerator-backed CI.
• Partner with model, compiler, kernel, runtime, firmware, validation, and hardware teams to translate release risks into automated gates.
• Improve CI reliability, speed, debuggability, and cost efficiency.

Required qualifications

• Built or operated large-scale CI/CD, developer infrastructure, test automation, or production engineering systems.
• Understand modern software delivery practices, including GitOps, release promotion, rollback, reproducibility, and policy-driven automation.
• Experience designing monorepos, build systems, dependency graphs, test selection, or change-impact analysis.
• Can design regression and performance-benchmarking systems with stable baselines, useful metrics, and actionable failure diagnosis.
• Proficient in Python, Go, Rust, C++, or another language used to build reliable infrastructure and automation.
• Have worked with distributed systems, schedulers, containers, clusters, or heterogeneous compute infrastructure.
• Can collaborate with model and systems engineers to turn complex accelerator workloads into repeatable production qualification.

About the company

OpenAI is an AI research and deployment company dedicated to ensuring that general-purpose artificial intelligence benefits all of humanity. The Hardware organization develops AI-native silicon and system-level solutions for advanced AI workloads.