Director, Engineering
Crusoe · San Francisco, CA - US
About this role
## Director, Engineering — GPU & Data Center Infrastructure Crusoe is on a mission to accelerate the abundance of energy and intelligence. We’re building AI infrastructure that’s faster for people innovating with AI—and better for the world—by owning and operating each layer of the stack (from electrons to tokens). We’re seeking a seasoned engineering leader to own and scale the GPU and data center infrastructure powering our neocloud platform. This is a hands-on leadership role at the intersection of software and hardware, with direct impact on reliability, automation, and revenue. --- ## About the Role You’ll lead teams building validation platforms, lifecycle automation, and orchestration services that keep our fleet healthy—while also owning firmware qualification, hardware security, deployments, and data center operations across a fleet growing ~10x year over year. Our fleet runs under a **99.5% monthly availability SLA**, so your reliability and automation work is directly tied to revenue. You’ll set the technical charter for fleet reliability and operational excellence, define north-star availability metrics, and build automation that replaces manual processes. You’ll partner closely with Engineering, Product, Finance, and the executive team. --- ## What You’ll Be Working On - **Infrastructure Deployment & Fleet Management**: Lead day-to-day fleet operations and greenfield deployments across co-location and owned facilities (from rack landing to customer availability). Own core fleet-health metrics and drive toward a **30-day time-to-sellable** target. - **Firmware Lifecycle Management**: Build the pipeline to qualify and track every firmware release across **10+ SKUs** and multiple OEMs/ODMs. Eliminate qualification backlog (e.g., ~**66-day NVIDIA firmware lag**), hold partners accountable, and evaluate in-house BMC development (e.g., **OpenBMC**). - **Validation Framework & Platform**: Consolidate siloed hardware automation pipelines and validation scripts into a single platform built on a workflow engine. Own bare-metal and multi-node burn-in/stress testing, define pass/fail criteria, deliver performance baselines (e.g., **MFU**) at cluster handover, and require stack-validation approval for production changes. - **Data Center Infrastructure Engineering (DCIE)**: Own power, cooling, networking fabrics, and physical security. Prioritize RMAs ahead of warranty expiry, reduce manual node recoveries via lifecycle automation, and make automation the source of truth for site deployments. Lead asset-visibility initiatives such as fleet-wide **RFID** and advanced **DCIM** tooling. - **Hardware Security & Attestation**: Own the strategy for trusted, tamper-evident hardware. Lead rollout of firmware/OS integrity attestation (TPM-based identity, secure boot gating, remote attestation) to quarantine compromised hardware—without impacting customer training or inference workloads. - **Reliability & Incident Operations**: Define operational KPIs (**uptime, PUE, MTTR, MTTD, MTBF**), establish runbooks, a tiered on-call model, and single-owner incident closure. - **Procurement, CapEx & Budget Ownership**: Partner with Procurement, Hardware, and Supply Chain on GPU and hardware sourcing (pricing, supply commitments, SLAs). Work with Hardware leadership on fleet capital and operating budgets, and report fleet health/spend/forecast to executives. - **Team Leadership & Outcome Alignment**: Build, develop, and retain a high-perform
Listing freshness
CronJobs last confirmed this listing 1h ago. If its source stops confirming the opening for seven days, this page is removed from active inventory.