Member of Technical Staff, Exceptional Generalist (Remote)
Core
Build and optimize the vLLM inference engine across the full stack, from low-level GPU kernels to high-level distributed systems, to make AI inference cheaper and faster.
Role type
Senior IC systems/ML infrastructure engineer (generalist)
Builds
vLLM inference runtime, GPU kernels, distributed orchestration systems, and cloud orchestration infrastructure
Domain
AI/ML infrastructure, GPU computing, distributed systems
Deliverable
production ML models
Required skills
CUDA kernels, Rust, Go, C++, Python, PyTorch, Kubernetes, Transformer architectures, KV-cache management
Preferred skills
Open-source contributions, multi-accelerator experience, quantization techniques, compiler technologies
Technologies
vLLM, TensorRT-LLM, SGLang, Triton, TileLang, Pallas, NVIDIA, AMD, TPU, Intel
Responsibilities
Optimize CUDA kernels for diverse hardware, design distributed orchestration systems, implement new model architectures, build cluster management and deployment automation (via careerplan.io/jobs/f0d2619d-28e0-4b25-8d30-3ac555071abb-member-of-technical-staff-exceptional-generalist-remote-at-inferact)
Seniority
Senior, autonomous IC
