XPU Architect

Who Are We? Postman is the world’s leading API platform, used by more than 45 million+ developers and 500,000 organizations, including 98% of the Fortune 500. Postman is helping developers and professionals across the globe build the API-first world by simplifying each step of the API lifecycle and streamlining collaboration—enabling users to create better APIs, faster. The company is headquartered in San Francisco and has offices in Boston, New York, Austin, Tokyo, London, and Bangalore - where Postman was founded. Postman is privately held, with funding from Battery Ventures, BOND, Coatue, CRV, Insight Partners, and Nexus Venture Partners. Learn more at postman.com or connect with Postman on X via @getpostman. P.S: We highly recommend reading The "API-First World" graphic novel to understand the bigger picture and our vision at Postman. The Opportunity As a Member of Technical Staff on AI Infrastructure, you will build and maintain the foundational systems and distributed infrastructure that power AI model post training, inference, and data pipelines. You will collaborate with engineering and research teams to ensure performance, scalability, and reliability of critical AI systems. What You’ll Do Design and implement large-scale, distributed AI infrastructure and services Optimize performance for GPU/xPU accelerators and cloud environments Build tools for observability, reliability, and scaling of AI workloads Partner with cross-functional teams to define AI infrastructure requirements and roadmap Contribute to architectural design and system longevity About You Have experience with GenAI infrastructure systems, distributed systems, cloud computing, and high-performance infrastructure Are proficient in programming languages like Python, Go, or similar Understand scaling challenges specific to AI workloads and accelerators Thrive in fast-paced, collaborative engineering environments The reasonably estimated base salary for this role ranges from $256,000.00 to $276,

About the Team The Core Network Engineering team owns the end-to-end networking stack that connects OpenAI’s compute infrastructure — spanning global WAN/edge connectivity, data-center networking, and high-performance host/xPU networking used for large-scale training and inference workloads. This team is responsible for ensuring networking is never the bottleneck to model training efficiency, cluster reliability, or fleet expansion. They design and operate the systems that provide predictable, high-throughput, low-latency connectivity across some of the world’s most advanced AI infrastructure. About the Role We’re looking for engineers to help build and operate the networking foundation behind OpenAI’s frontier AI systems. Depending on your background and area of focus, you may work across host networking, datacenter fabrics, or global WAN infrastructure. The problems span low-level systems software, distributed infrastructure, protocol readiness, observability, performance engineering, automation, and large-scale network operations. You’ll work on systems where microseconds of latency, tail performance, and network reliability directly impact model training efficiency and production serving performance. This role is ideal for engineers who enjoy operating close to the hardware/software boundary and solving performance-critical infrastructure problems at massive scale. In this role, you will: Design, build, and operate networking systems that support large-scale AI training and inference infrastructure Improve performance, reliability, and scalability across host networking, datacenter fabrics, and WAN systems Develop automation for provisioning, configuration management, validation, upgrades, and lifecycle management of networking infrastructure Build tooling and observability systems for network health, performance analysis, debugging, and automated remediation Optimize network performance across technologies such as RDMA, RoCE, InfiniBand, Ethernet, and high-perf

About the Team OpenAI’s 1P Hardware Systems team operates at the intersection of Hardware Engineering, Infrastructure, Supply Chain, Manufacturing, Deployment, and Finance to translate infrastructure demand into an executable systems plan. The Planning Lead connects system demand and deployment timing to site and configuration requirements, power availability, XPU needs, hardware supply commitments, manufacturing capacity, and site readiness. About the Role OpenAI is seeking a 1P Hardware Systems Planning Lead to own the integrated demand, supply, and deployment plan for our 1P AI infrastructure systems. This role will connect infrastructure demand and site power availability to system configurations, XPU requirements, and hardware supply commitments—creating a single, actionable view of whether our deployment plan can be met. You will establish the planning mechanisms that allow teams to see what changed, understand the impact, and act quickly when demand, supply, configuration, site readiness, or power timing moves. You will work across Hardware Engineering, Infrastructure, Supply Chain, Strategic Sourcing, Manufacturing, Deployment, and Finance to identify gaps early and drive recovery plans to closure. This is a highly cross-functional role for someone who combines systems-level thinking, strong analytical judgment, and rigorous program execution. The ideal candidate can turn complex and changing inputs into a clear operating plan, surface the decisions that matter, and drive accountability across teams without relying on formal authority. Success also requires strong communication judgment: the ability to align cross-functional teams, brief leadership at a concise and decision-oriented level, and go deep into the underlying assumptions, dependencies, risks, and recovery plans when needed. Responsibilities Integrated Planning - Own the integrated planning process for 1P hardware systems, connecting system demand, deployment timing, site and configuration require