Principal AI Cluster Performance Validation Engineer
Core
Optimize and validate peak performance for GPU clusters, specifically focusing on RDMA networks, data flows between GPU/NIC, and cluster networking.
Role type
Principal AI Cluster Performance Validation Engineer
Builds
High-performance GPU cluster infrastructure for AI workloads
Domain
Semiconductor / High-Performance Computing (HPC) / AI Infrastructure
Deliverable
production ML models
Required skills
GPU architectures, RDMA networking, RoCE v2, parallel clustered computing, system-level performance tuning, Linux kernel networking, scripting (Python/Bash), performance profiling, bottleneck analysis
Preferred skills
Machine learning system design, cluster management tools, hardware/firmware debugging, load balancing techniques
Technologies
RDMA, RoCE, Linux, Python, Bash
Responsibilities
Conduct scalability testing under various workloads; Develop and execute benchmarking strategies to identify bottlenecks; Optimize cluster network throughput, latency, and congestion; Profile performance to provide actionable insights; Implement optimization strategies for protocols and parallel processing; Document performance analysis and tuning outcomes; Collaborate with cross-functional teams to integrate improvements.
Seniority
Principal, strategy & mentorship
