Compute Platform Engineer

Responsibilities

• Designing, configuring, and managing a High performance compute infrastructure made up of GPU and CPU nodes
• Manage the full firmware/BIOS lifecycle across our HPC/AI fleet – from baselines and validation through rollout and compliance
• Troubleshoot hardware components (CPU, GPU, DPU, NVSwitch, NICs, memory, PSU, BMC) and guide replacement or configuration changes
• Diagnose and automate recurring hardware issues to improve reliability and reduce recovery time
• Work on the latest AI platforms from day one (e.g., NVL72 / Grace Blackwell), ensuring they are stable, performant, and ready for production use
• Monitoring hardware performance, identifying areas for improvement, and implementing solutions
• Automate health checks and onboarding workflows to accelerate safe deployment
• Collaborate with vendors on firmware issues – providing clear repro cases, logs, and impact to drive fixes and improvements
• Recommend process, tooling, and architectural improvements to strengthen platform operations
• Performing diagnostics, tuning, and capacity planning to ensure smooth scale-out
• Performing analysis of existing hardware lifecycle processes and providing recommendations for improvement and optimization
• Collaborating with various teams to integrate hardware improvements and align with organizational goals
• Implementing best practices for security hardening of the platform and associated systems
• Mentoring junior engineers and fostering a culture of continuous learning and improvement
• Acting as a subject matter expert, providing guidance and support for infrastructure-related issues
• Leveraging Infrastructure as Code (IaC) methodologies to ensure efficient and scalable infrastructure management

Requirements

• 3+ years of hands-on experience supporting large-scale compute platforms
• Proficiency with HPE server infrastructure, such as ProLiant and Apollo, and NVIDIA GPUs, including A100 and H200
• Solid understanding of server architecture, including UEFI/BIOS, PCIe devices and out-of-band management systems, such as iLO and BMC
• Proven ability to resolve complex hardware issues and manage vendor relationships
• Familiarity with automation tools such as Ansible, Terraform and CI/CD systems
• Working knowledge of Linux in high-performance or latency-sensitive environments
• Working knowledge of basic network concepts, such as DNS, DHCP, VLANs, switching and routing
• Basic working knowledge of Kubernetes and Openstack technologies (preferred but not required)
• Experience with data center operations and process adherence
• Excellent communication and coordination skills with cross-functional teams and external partners