Member of Technical Staff – Software Engineer, GPU Cluster Infrastructure

Perplexity

GPU infrastructure engineer building Perplexity's self-serve compute platform for AI training and inference. Operating multi-cloud Kubernetes clusters, scheduling scarce GPU capacity, and improving reliability.

Posted 8/5/2026full-timeSan Francisco • California • 🇺🇸 United StatesLead💰 $250,000 - $485,000 per yearWebsite

Core Competencies

Role fit

Core Competencies

Use this summary to align your resume positioning with the role.

Demonstrates expertise in building and managing GPU cluster infrastructure, with a strong focus on Kubernetes, distributed systems, and high-availability services. Proficient in orchestrating compute across multiple cloud providers while ensuring workload resilience and observability.

Highest-signal resume keywords

Deep Kubernetes ExperienceGPU Cluster ManagementInfrastructure Code in Go, Rust, or C++Distributed Systems FundamentalsObservability for ML Workloads

ATS Keywords

Tailor your resume

Applicant Tracking System Keywords

Tip: use these terms in your resume and cover letter to boost ATS matches.

Hard Skills

Kubernetes OperatorsCustom Resource Definitions (CRDs)NVIDIA Hardware KnowledgeCUDAHigh-Speed NetworkingMulti-Cloud OrchestrationScheduling and Resource AllocationFault ToleranceLong-Running Training JobsInference Serving Stacks

Soft Skills

End-to-End OwnershipProblem-Solving

Tools & Technologies

PrometheusGrafanaWeights & BiasesSlurmInfiniBandRoCERDMA

Industry Keywords

GPU Fleet ProvisioningCompute PlatformCloud ProvidersWorkload ResilienceCapacity Integration

Tech Stack

Tools & technologies

AWSCloudDistributed SystemsGoGoogle Cloud PlatformGrafanaKubernetesPrometheusRust

About the role

Key responsibilities & impact

Build and own a self-serve compute platform for launching training jobs and operating inference services without direct GPU provisioning or cluster management
Operate GPU fleet provisioning, lifecycle management, reliability, and capacity integration across cloud providers
Build scheduling and placement logic to find and efficiently pack available GPU capacity and assign workloads to suitable hardware
Support long-running distributed training jobs and highly available, low-latency production inference services on the same fleet
Write Kubernetes operators and CRDs and manage multi-provider GPU clusters
Build fault tolerance, autoscaling, and observability for workload resilience and fleet utilization
Partner with inference and cloud infrastructure engineers on platform architecture and roadmap
Take end-to-end ownership of GPU cluster infrastructure at Perplexity, which serves hundreds of millions of queries monthly

Requirements

What you’ll need

Deep Kubernetes experience, including custom operators, CRDs, and multi-cluster federation
Experience managing GPU clusters at scale
Knowledge of NVIDIA hardware, CUDA, and high-speed networking such as InfiniBand or RoCE
Experience orchestrating compute across multiple clouds, including CoreWeave, AWS, GCP, or similar
Strong distributed systems fundamentals in scheduling, resource allocation, and fault tolerance under load
Ability to write infrastructure and systems-level code in Go, Rust, or C++
Experience supporting long-running training jobs and high-availability inference services
Experience with inference serving stacks such as vLLM, SGLang, or TensorRT-LLM is valued
Experience with Slurm or other HPC schedulers is valued
GPU kernel experience in CUDA or Triton is valued but not required
Production experience with InfiniBand, RoCE, or RDMA is valued
Observability experience for ML workloads with Prometheus, Grafana, or Weights & Biases is valued
Ability to own problems end-to-end in an environment without a predetermined path

Benefits

Comp & perks

Equity
Health insurance
Dental insurance
Vision insurance
Retirement benefits
Fitness benefits
Commuter benefits
Dependent care accounts
Regionally tailored benefits for full-time employees outside the U.S.