Senior/Principal Product Manager - Cluster Platform
Core
Own the product experience and operational lifecycle for bringing new AI infrastructure clusters onto Radiant's platform, making deployment repeatable, measurable, automated, and production-ready at scale.
Role type
Senior/Principal Product Manager (AI Infrastructure)
Builds
AI-native infrastructure platforms for GPU-native workloads and multi-tenant control planes
Domain
AI Infrastructure / High-Performance Computing
Deliverable
production ML models
Required skills
Cluster systems thinking, Product mindset, Technical fluency (Compute, Networking, Storage, Orchestration, Automation, Observability), Operational discipline, Automation mindset, Ownership
Preferred skills
Familiarity with benchmarking and validation tools (NCCL, HPL, MLPerf), Strong prioritisation, Communication, Stakeholder management
Technologies
Kubernetes, Slurm, Containers, RDMA, InfiniBand, Ethernet, Infrastructure-as-code, Configuration management, Workflow engines, NCCL tests, HPL, MLPerf
Responsibilities
Own strategy, roadmap, and delivery for cluster onboarding, validation, acceptance, and lifecycle management; Define end-to-end process from installed hardware to production-ready cluster; Productise and automate workflows across discovery, provisioning, configuration, testing, benchmarking, acceptance, and handover; Define cluster readiness and acceptance criteria across compute, GPUs, networking, storage, firmware, software, and reliability; Partner with engineering to build tooling and automation for cluster deployment, validation, remediation, and lifecycle operations; Define benchmarking frameworks for GPU performance, network fabric, storage throughput, collective communication, and workload performance; Build clear visibility into cluster state, readiness, failures, exceptions, and deployment progress; Identify recurring deployment friction and turn operational processes into scalable platform capabilities; Drive prioritisation, technical requirements, delivery, adoption, and continuous improvement across the cluster lifecycle; Define success metrics across time-to-ready, automation coverage, acceptance pass rates, deployment reliability, and cluster performance consistency.
