Senior Systems Software Engineer, Workload Systems Engineering
Core
Build self-service mechanisms, diagnostics, and validation workflows to enable engineering teams to run demanding build, simulation, and validation workloads efficiently and reliably across distributed infrastructure.
Role type
Senior Systems Software Engineer (Workload Systems)
Builds
Self-service configuration tools, diagnostic views, and infrastructure improvements for large-scale build and test systems.
Domain
High-performance computing, distributed systems, GPU infrastructure, and CI/CD pipelines.
Required skills
Systems programming, Linux, distributed systems, performance profiling, debugging complex reliability issues, GPU systems/drivers, scheduling and resource management, remote execution, storage and data paths.
Preferred skills
GPU sharing and isolation (CUDA, MIG, MPS), workload placement, NUMA, cloud-instance benchmarking, Bazel, simulation/replay workloads, hardware-in-the-loop validation.
Technologies
Linux, CUDA, MIG, MPS, Bazel, CI/CD tools, profiling tools, metrics, traces.
Responsibilities
Develop shared self-service and validation mechanisms for infrastructure workflows; provide diagnostic views across application, submission, and host layers; set performance and reliability benchmarks using profiling and experiments; evaluate and improve compute/GPU selection, scheduling, and resource sharing; implement infrastructure-level changes and measure impact on efficiency and cost; turn investigations into supported workflows and documentation.
Seniority
Senior, hands-on IC
