Build and own a self-service compute platform for training and inference workloads.
Provision, manage, and operate GPU fleets across multiple cloud providers.
Develop scheduling and placement logic to efficiently allocate scarce GPU capacity.
Support both long-running distributed training jobs and highly available production inference services.
Write Kubernetes operators and CRDs and manage multi-cluster GPU orchestration.
Build fault tolerance, autoscaling, and observability to handle node loss, provider failures, and capacity changes.
Set technical direction and develop platform architecture and roadmap with inference and cloud infrastructure teams.
Requirements
Deep Kubernetes experience, including custom operators, CRDs, and multi-cluster federation.
Experience managing GPU clusters at scale with NVIDIA hardware, CUDA, and InfiniBand or RoCE networking.
Experience orchestrating compute across multiple clouds such as CoreWeave, AWS, or GCP.
Strong distributed systems fundamentals in scheduling, resource allocation, and fault tolerance.
Ability to write infrastructure and systems-level code in Go, Rust, or C++.
Experience supporting both long-running training jobs and high-availability inference services.
Additional valued experience includes vLLM, SGLang, TensorRT-LLM, Slurm, GPU kernel development with CUDA or Triton, InfiniBand, RoCE, RDMA, Prometheus, Grafana, or Weights & Biases.
Salary: $250k - $485k/yr
Perplexity
The most powerful answer engine. Powering curiosity with answers backed by up-to-date sources. This is where knowledge begins.