Staff AI Performance Engineer
Core
Lead performance work to optimize AI training and inference workloads across distributed systems by connecting compute, memory, networking, and software behavior.
Role type
Staff AI Performance Engineer
Builds
Benchmarks, profiling tools, and optimization strategies for large-scale AI infrastructure
Domain
AI Infrastructure / High-Performance Computing / Data Center Systems
Deliverable
production ML models
Required skills
Profiling and optimizing AI/ML/HPC workloads, Distributed systems (MPI, NCCL, UCX, libfabric), C++, Python, Compute/memory/communication behavior in large-scale systems, Technical leadership and cross-team coordination
Preferred skills
MLPerf familiarity, Accelerated architectures, ML frameworks, High-performance interconnects
Technologies
MPI, NCCL, UCX, libfabric, C++, Python
Responsibilities
Analyze and optimize AI training and inference workloads, Build benchmarks and profile demanding workloads, Develop C++ and Python tools for analysis and optimization, Lead complex performance workstreams across hardware and software, Find bottlenecks and validate improvements, Coordinate improvements across teams
Seniority
Staff, technical leadership & cross-team coordination
