Platform Engineer: Self-Serve GPU Clusters & Kubernetes

Perplexity; United States; Remote; USD 180,000 - 320,000; Full time

Job summary

Perplexity is building a self-serve platform that hides infrastructure complexity for training and inference workloads. You will design and own systems enabling researchers to launch training jobs and run inference services without manual GPU provisioning or provider-specific configs.

You’ll manage a GPU fleet across multiple clouds, implement scheduling and placement that respects real constraints, and drive reliability, observability, and scale across clusters.

Qualifications

• Deep Kubernetes experience with custom operators and multi-cluster federation.
• Experience managing GPU clusters at scale with NVIDIA hardware and CUDA networking.
• Orchestrated compute across multiple clouds (CoreWeave, AWS, GCP) and understood differences between them.
• Strong distributed systems fundamentals: scheduling, resource allocation, fault tolerance under load.
• Proficient in Go, Rust, or C++ for infrastructure and systems programming.
• Experience supporting both long-running training jobs and high-availability inference workloads.
• Own problems end-to-end and excel when the path forward isn't laid out.

Responsibilities

• Build a self-serve compute platform. Design and own the systems that let inference engineers and researchers launch training jobs and operate inference services without managing GPU provisioning, cluster configuration, or provider-specific infrastructure.
• Operate the GPU fleet. Own provisioning, lifecycle management, reliability, and capacity integration across providers, giving teams a consistent way to use compute regardless of where it runs.
• Solve for GPU scarcity. Build the scheduling and placement logic that finds available capacity across providers, packs it efficiently, and gets the right workload onto the right hardware under real constraints.
• Support two very different workloads. Keep long-running distributed training jobs healthy while simultaneously guaranteeing the availability and latency of production inference services on the same fleet.
• Own the Kubernetes for GPU orchestration. Write the operators and CRDs, and manage many clusters across providers so the platform behaves the same everywhere we run.
• Make failure boring. Build the fault tolerance, autoscaling, and observability that keep the fleet utilized and let workloads survive node loss, provider hiccups, and capacity shifts without human intervention.
• Set technical direction across teams. Partner with inference and cloud infrastructure engineers to turn operational constraints into a coherent platform architecture and roadmap.

Skills

Kubernetes experts

GPU clusters at scale

Multi-cloud orchestration

Distributed systems fundamentals

Go/Rust/C++

Training & inference workloads

End-to-end ownership

Job description

Perplexity is building a self-serve platform that hides infrastructure complexity for training and inference workloads. You will design and own systems enabling researchers to launch training jobs and run inference services without manual GPU provisioning or provider-specific configs.

You’ll manage a GPU fleet across multiple clouds, implement scheduling and placement that respects real constraints, and drive reliability, observability, and scale across clusters.