Member of Technical Staff (Software Engineer, GPU Cluster Infrastructure)
Core
Build a unified, self-serve platform to manage GPU clusters for AI training and inference workloads across multiple cloud providers.
Role type
Senior IC GPU cluster infrastructure engineer
Builds
Self-serve compute platform for training and inference
Domain
Cloud infrastructure, GPU orchestration, distributed systems
Deliverable
production ML models | infrastructure
Required skills
Kubernetes (operators, CRDs, multi-cluster), GPU cluster management (NVIDIA, CUDA), multi-cloud orchestration, distributed systems (scheduling, fault tolerance), systems programming (Go, Rust, C++)
Preferred skills
Inference serving stacks (vLLM, SGLang, TensorRT-LLM), HPC schedulers (Slurm), GPU kernel development (CUDA, Triton), high-speed interconnects (InfiniBand, RoCE, RDMA), ML observability (Prometheus, Grafana, Weights & Biases)
Technologies
Kubernetes, NVIDIA, CUDA, InfiniBand, RoCE, Go, Rust, C++, CoreWeave, AWS, GCP, vLLM, SGLang, TensorRT-LLM, Slurm, Prometheus, Grafana, Weights & Biases
Responsibilities
Design and own systems for launching training jobs and operating inference services; manage GPU fleet provisioning, lifecycle, and capacity across providers; build scheduling logic for efficient GPU utilization; maintain health of distributed training jobs and production inference services; write operators and CRDs for Kubernetes GPU orchestration; implement fault tolerance, autoscaling, and observability; define technical direction for platform architecture
Seniority
Senior, hands-on IC
