What you'll be doing:

Characterize node, rack, and cluster performance using microbenchmarks and representative applications to identify compute, memory, communication, and storage bottlenecks.
Identify Linux performance problems related to CPU scheduling, NUMA locality, CPU and memory affinity, interrupt handling, and resource contention and assess their impact on GPU usage and application efficiency.
Examine distributed workload scaling considering communication, synchronization delays, slow nodes, and the interplay between computation and communication, including GPU and NIC positioning, interconnect structure, and network traffic impact.
Combine application profiles, system traces, hardware counters, and cluster telemetry to isolate root causes; develop reproductions and collaborate with hardware, firmware, driver, library, and application teams for fixes.
Develop and maintain Python tools for automated benchmark execution, data collection, and performance analysis, setting up baselines, assessing variability, and spotting regressions.
Define and complete performance validation for new platforms and cluster deployments, identifying underperforming components and verifying corrective actions.

Requirements

Extensive practical experience in Linux performance analysis, including processes and threads, CPU and memory architecture, NUMA, I/O, and networking.
Experience with performance investigation in multi-node HPC or GPU-accelerated systems, differentiating local bottlenecks from communication overhead and resource contention.
Deep expertise in at least one relevant area: GPU application performance, Linux systems performance, high-speed networking and collective communication, parallel storage, or HPC performance tooling.
Proficiency in Python for automation and analysis; experience in developing and debugging performance-sensitive code in C, C++, and CUDA.
Ability to build controlled experiments, interpret profiling and counter data, and demonstrate reproducible performance improvements.
Strong independent technical initiative across hardware/software boundaries with clear communication and collaboration skills.
BS or MS in Computer Science, Engineering, Mathematics, Physics, or equivalent experience with 8+ years of applicable experience.

Experience with profiling and tracing tools such as Nsight Systems, Nsight Compute, Linux perf, or eBPF-based tools.
Practical experience with RDMA, MPI or NCCL, and analyzing performance across PCIe, NVLink, InfiniBand, or Ethernet interconnects.
Experience with Slurm, container runtimes, and cgroups, and understanding their impact on workload placement and resource isolation.
Experience improving distributed training, inference, or scientific applications via scaling studies or automated performance regression frameworks like ReFrame.