Platform Engineer, GPU Inference & Training Orchestration

Perplexity is seeking an experienced platform engineer to own a unified, self-serve compute platform for training and inference workloads. You will design systems that launch training jobs and operate inference services without GPU provisioning burdens. You will manage the GPU fleet across providers, ensure reliability, and build scheduling that fits capacity constraints while supporting both long-running training and high-availability inference workloads. #J-18808-Ljbffr