Overview

OCI AI Infrastructure leads in building cutting-edge GPU supercomputers scaling to tens of thousands of GPUs without performance loss. Our team are experts in RDMA cluster architecture and its impact on AI/ML/HPC performance. We leverage this expertise to optimize RDMA network designs, enabling customers to advance AI/ML and HPC frontiers.

Responsibilities

• Design and execute performance benchmarks for RDMA-based GPU clusters at massive scale.
• Analyze network bottlenecks in AI/ML/HPC workloads using RDMA fabrics.
• Collaborate with hardware engineers to optimize cluster networking architectures.
• Develop tools and methodologies for measuring end-to-end cluster performance.
• Drive performance tuning for multi-tenant GPU supercomputer environments.
• Research emerging RDMA technologies and their AI/HPC implications.
• Create performance models predicting scalability for 10k+ GPU clusters.
• Document findings and present recommendations to stakeholders and customers.

Qualifications

• BS/MS in Computer Science, Electrical Engineering, or related field.
• 5+ years experience in high-performance networking or HPC systems.
• Deep knowledge of RDMA (RoCE, IB) and cluster interconnects.
• Strong experience with GPU computing and AI/ML workloads.
• Proficiency in performance analysis tools (e.g., perf, nsight, ibv tools).
• Excellent programming skills in C/C++, Python for benchmarking.

Benefits

• Competitive salary and equity package.
• Comprehensive health, dental, and vision insurance.
• 401(k) matching and employee stock purchase plan.
• Flexible PTO and remote/hybrid work options.
• Professional development budget and conference sponsorships.