Senior Performance Engineer
Core
Build and evolve systems for performance analysis, telemetry, and optimization of large-scale GPU- and CPU-based clusters for AI and HPC.
Role type
Senior IC performance engineer (systems/HPC)
Builds
Performance analysis, benchmarking, and diagnostic tools for AI/HPC workloads
Domain
High-performance computing (HPC) and AI infrastructure
Deliverable
production ML models
Required skills
Performance analysis, high-performance networking (RDMA, MPI, NCCL), system architecture, telemetry collection, Python, Linux
Preferred skills
CUDA internals, congestion control algorithms, CPU/GPU architecture, PyTorch, TensorFlow, Bash, C/C++
Technologies
NCCL, RDMA, MPI, RoCE, CUDA, PyTorch, TensorFlow, Python, Bash, C/C++, Linux
Responsibilities
Profile and benchmark AI/HPC workloads on GPU/CPU clusters, analyze high-performance networking characteristics, identify bottlenecks across networking/compute/memory, develop performance tools, define test plans, collaborate with hardware/firmware/software teams, support telemetry collection
Seniority
Senior, hands-on IC
