AI Platform Architect

About the Company

Graphcore is one of the world’s leading innovators in Artificial Intelligence compute. As part of the SoftBank Group, Graphcore is developing hardware, software and systems infrastructure that will unlock the next generation of AI breakthroughs and power the widespread adoption of AI solutions across every industry.

Responsibilities

• Define the holistic architecture for highly clustered AI environments to ensure zero-bottleneck data flow
• Influence strategy for AI workload scheduling and orchestration using tools like Kubernetes or Slurm
• Profile and eliminate system-level bottlenecks across the entire AI pipeline, from deep learning frameworks to OS-level tuning
• Work closely with software, firmware, and OS engineering to influence platform design and hardware exploitation
• Drive the 3-to-5-year technical vision for the AI platform and collaborate with subject matter experts to define requirements for silicon teams

Requirements

• 4+ years of experience as a Lead or Principal Architect for large-scale AI or machine learning platforms
• Deep practical knowledge of distributed AI training and deployment (data/tensor/pipeline parallelism)
• Authoritative understanding of PCIe Gen 5/6, NVMe namespaces, and RDMA (RoCEv2/InfiniBand)
• Experience with container orchestration and infrastructure-as-code (IaC) for GPU-heavy environments
• Ability to bridge the gap between AI researchers and low-level hardware engineers

Preferred Qualifications

• Hands-on experience with rack-scale GPU AI platforms
• Experience with AI platforms integrating latest networking, cooling, and GPU technologies
• Working knowledge of Python or JSON for workload characterization