Inference & Workload Optimization Lead (Horizon AI)
Core
Lead the optimization of customer AI training and inference workloads on dedicated GPU clusters to improve speed and efficiency.
Role type
Principal GPU performance engineering lead (inference & workload optimization)
Builds
Autonomous agents that profile, tune, and optimize customer workloads; cost/latency-aware internal model serving routing.
Domain
AI Infrastructure / GPU Performance Engineering / Data Center Operations
Deliverable
production ML models
Required skills
GPU performance engineering, workload profiling, distributed training and inference, collective communication tuning, serving stack expertise, memory and bandwidth optimization, kernel-level understanding, performance baselining, agent-encodable optimization logic.
Preferred skills
CUDA kernel development, open-weight model deployment at scale, benchmark design, customer-facing technical experience.
Technologies
NVIDIA GPUs, vLLM, SGLang, TensorRT-LLM, NCCL, Nsight, DCGM.
Responsibilities
Profile customer training and inference workloads to identify bottlenecks; establish performance baselines per hardware generation; encode optimization expertise into autonomous agents; tune workloads for committed-capacity customers; manage internal model serving routing and version migration; advise on GPU allocation between internal and customer workloads.
Seniority
Principal, hands-on IC with strategic scope.
