SW Engineer, Hardware and System Bring-up/Bootstrapping - Industrial Compute, Scaling

SW Engineer, Hardware and System Bring-up/Bootstrapping - Industrial Compute, Scaling About the Team The Scaling team is responsible for the architectural and engineering backbone of OpenAI’s infrastructure. We design and deliver advanced systems that support the deployment and operation of cutting-edge AI models. Our work spans system software, networking, platform architecture, fleet-level monitoring, and performance optimization. About the Role We’re looking for an engineer who can take early, sometimes messy, pre-production hardware and make it “real”: bootstrapped, stable, imaged, joined to the right Kubernetes control plane, registered correctly, scheduled, and observable. You’ll sit at the intersection of early HW bring-up, provisioning automation, fleet/cluster management systems, and lab or cloud provider integration—turning new SKUs into capacity that is usable by internal customers. Key Responsibilities - Own the end-to-end bring-up and bootstrap path for new systems and compute nodes from bare metal/early access in lab or production/cloud environments to schedulable fleet capacity: image build, user-data/config, cluster join, and readiness gates. - Build and maintain “first-class” golden image + provisioning workflows across lab, and production environments, including working with partner-provided base images and reconciling OS/version requirements. - Work with partner teams to integrate nodes into our fleet infrastructure and IaC pipelines (Terraform, Chef, etc.), ensuring cloud resources map cleanly onto our internal lifecycle expectations (e.g., VMSS/instance pools, image references). - Partner with scheduling and platform owners to ensure new hardware is reachable and scheduled (pool definitions, network/WAN connectivity/routing, admission controls, platform-specific quirks), including cases where new SKUs require changes for scheduling integration. - Drive registration and inventory correctness (e.g., systems that track nodes and their metadata), including hands-on support to get nodes registered and visible end-to-end. - Collaborate with partner teams to implement baseline health + telemetry bring-up: minimum viable health signals, pass/fail checks, and automated reporting suitable for early ramp decisions. - Debug issues across layers: PXE/boot-loader, UEFI/BIOS, BMC, OS bring-up, NIC/network reachability, kubelet/control-plane connectivity, storage constraints, and early rack/lab realities. Qualifications - BS in CS/EE (or equivalent practical experience). - 5+ years of experience in systems SW development and building/operating Linux-based infrastructure in production or pre-production environments. - Strong, hands-on experience with: - Kubernetes cluster operations (node lifecycle, bootstrap/join, debugging control-plane connectivity) - Infrastructure-as-Code / config management (Terraform, Chef/Ansible, etc.) - Provisioning and imaging (PXE/iPXE, golden images, cloud-init/user-data) - Networking fundamentals (L2/L3, routing, DNS, firewalling; comfort debugging reachability) - Proven ability to write automation in Python/Go/Bash and ship operational tooling/runbooks. Preferred Skills - Experience bringing up new hardware platforms (early silicon/servers/NICs) in a lab setting and turning them into stable fleet capacity. - Multi-cloud operational experience (Azure/GCP/AWS/OCI), especially with compute pools (e.g., VMSS / instance pools). - Experience building telemetry/health pipelines (agent-based metrics/logging, health rollups, readiness criteria). - Familiarity with WAN, peering, and multi-site network concepts for cluster deployments. About OpenAI OpenAI is an AI research and deployment company dedicated to ensuring that general-purpose artificial intelligence benefits all of humanity. We push the boundaries of the capabilities of AI systems and seek to safely deploy them to the world through our products. AI is an extremely powerful tool that must be created with safety and ... (truncated, view full listing at source)