Augminter
HPC AI Architect
office • SAU • senior • Не указана
Augminter is hiring a HPC AI Architect for a client engagement in Riyadh, Saudi Arabia (full-time, on-site).
Key Responsibilities and Tasks:
• Design end-to-end HPC and AI infrastructure architectures covering compute, GPU, CPU, storage, and management environments.
• Define HPC storage architecture, including parallel file systems, scale-out NFS, performance tiers, capacity, and data access requirements.
• Size compute and storage infrastructure based on HPC and AI workload requirements, performance targets, and projected demand.
• Define high-performance interconnect and infrastructure requirements to support distributed HPC and AI workloads.
• Develop architecture options and assess trade-offs across performance, scalability, availability, and cost.
• Review technical designs, vendor proposals, and proposed HPC and AI infrastructure solutions.
• Coordinate with infrastructure, network, data center, AI, and operations teams to ensure end-to-end architecture alignment.
• Develop and maintain high-quality architecture documentation, sizing models, technical specifications, and design recommendations.
Required Experience and Knowledge:
• 10+ years of relevant experience in HPC, AI infrastructure, or high-performance computing architecture.
• Fluency in English is required.
• Strong experience designing GPU/CPU clusters, HPC environments, and high-performance storage solutions.
• Strong knowledge of parallel file systems, scale-out NFS, high-performance storage, and distributed computing.
• Experience with HPC cluster management, high-speed interconnects, and large-scale compute environments.
• Proven experience working with HPC/AI infrastructure vendors and technology integrators.
Education and Certification:
• Bachelor's degree in Computer Engineering, Computer Science, Information Technology, or an equivalent discipline is required.
• Red Hat Certified Engineer (RHCE), Red Hat Certified System Administrator (RHCSA), and Certified Kubernetes Administrator (CKA) are a must to have.