Systems Generalist, GPT Infrastructure
Core
Design and operate an automated inference optimization platform that generates, compiles, executes, and grades candidate kernels and runtime configurations on diverse accelerator hardware.
Role type
Senior IC systems generalist (distributed systems & AI inference infrastructure)
Builds
Control planes, APIs, secure partner-side execution environments, evaluation systems, and artifact pipelines for AI inference optimization.
Domain
AI infrastructure, distributed systems, compiler technology, and accelerator hardware optimization.
Deliverable
production ML models | infrastructure
Required skills
C++, Python, Go, or Rust; distributed systems architecture; Linux and networking; job orchestration; performance profiling and benchmarking; complex system debugging.
Preferred skills
AI inference-serving systems; compilers and runtimes (LLVM, MLIR, Triton, CUDA, ROCm); hardware architecture and ISA concepts; inference-serving frameworks (vLLM, SGLang, Triton Inference Server); secure partner-facing infrastructure.
Technologies
C++, Python, Go, Rust, Linux, Kubernetes, Docker, LLVM, MLIR, Triton, CUDA, ROCm, vLLM, SGLang, Triton Inference Server.
Responsibilities
Design durable APIs and control-plane services for multi-day optimization campaigns; build secure partner-side runner and grader software; integrate hardware profiles and compilers into repeatable workflows; develop correctness and performance evaluation systems; build artifact and qualification workflows; collaborate with research and infrastructure teams to deliver production solutions.
Seniority
Senior, hands-on IC
