Description

In this role, you will focus on enhancing the reliability and performance of GPU clusters for machine learning. You will guide product direction from discovery to adoption while collaborating with engineering teams and conducting in-depth customer research.

Requirements

• •3–5+ years of experience in product management, ML infrastructure/MLOps, distributed systems engineering, or cloud architecture
• •Strong technical foundation in computer science, distributed systems, or ML infrastructure
• •Hands-on familiarity with ML training, ideally using orchestrators like Slurm, Kubernetes, Ray, or similar systems
• •Proven ability to ship technically complex features with multiple engineering teams
• •Excellent communicator capable of influencing engineering, research, and customer stakeholders
• •Experience with product analytics, data-driven prioritization, and experiment design
• •Strong willingness and ability to learn quickly in a fast-evolving ML and infrastructure environment
• •Experience working with GPU platforms, Infiniband/RDMA networking, or HPC systems
• •Understanding of modern ML frameworks (PyTorch, DeepSpeed, FSDP, NCCL, etc.)
• •Knowledge of ML training efficiency: Goodput, MFU, scheduling, health checks
• •Exposure to LLM training, distributed data/ZeRO/FSDP strategies, or transformer inference
• •Experience in observability, performance tuning, or reliability engineering
• •Customer-facing technical experience (supporting ML or infrastructure workloads)
• •Strong candidates with backgrounds in ML infrastructure, distributed systems, SRE, or cloud engineering who want to grow into product

Tasks

• •Own key tracks in Cluster Experience
• •Define product direction from problem discovery to delivery
• •Drive cross-functional execution across various teams
• •Conduct deep customer research through interviews and analytics
• •Identify bottlenecks across hardware, network, and runtime
• •Translate ML research ideas into scalable product features
• •Shape user interactions with clusters through dashboards and notifications

Tools & Technologies

Benefits