The posting, in OpenAI's own words
archived Oct 29, 2026About the Team
The Cloud Infrastructure team builds and operates the infrastructure behind OpenAI’s products, including ChatGPT and the API. Within Cloud Infrastructure, the Traffic team owns how traffic enters our cloud environments, moves between services, clusters, and regions, and reaches external dependencies. Our work spans cloud networking, Kubernetes networking, ingress and egress gateways, proxies, load balancing, and traffic steering. We build the systems, controls, and observability that make these paths reliable, secure, efficient, and straightforward for other engineering teams to use.
About the Role
You’ll build and operate the networking and traffic infrastructure that connects OpenAI’s products and services across clouds and regions. The team’s work spans L3, L4, and L7—from IP connectivity, NAT, and transport behavior to service proxies and application request routing. You’ll bring depth in part of this stack and curiosity about how the whole path works. Projects may include building gateways and control planes, improving cross-region connectivity, diagnosing packet loss or connection exhaustion, and making routing changes safer and easier to operate. This role combines software engineering with production ownership. You’ll turn ambiguous networking problems into durable improvements, working closely with service owners, infrastructure teams, Security, and cloud providers. This role is based in San Francisco, CA or Seattle, WA.
Read the full posting ↓
In this role, you will:
Design, build, and operate ingress, egress, and service-to-service connectivity across clusters, regions, and clouds. Develop proxies, gateways, controllers, and automation for routing, load balancing, failover, and traffic management. Build and evolve cloud networking infrastructure, including private connectivity, NAT, and network isolation. Diagnose failures and performance bottlenecks across application, proxy, transport, and cloud networking layers; work with cloud providers when issues extend into their infrastructure. Build telemetry and diagnostic tools that explain traffic paths, identify failing components, and help engineers take effective action. Partner with Security to implement and validate traffic-path protections, including secure service communication, access controls, egress restrictions, and DDoS defenses. Improve latency, throughput, availability, and network efficiency through measurement, testing, and careful production rollouts. Create clear interfaces and safe defaults for service teams, participate in on-call, and turn operational lessons into changes that prevent recurring failures and reduce manual work.