AI Platform Architect

Graphcore•Austin, TX

About The Position

Graphcore is seeking a visionary AI Platform Architect to design and oversee the infrastructure stack supporting demanding distributed AI workloads. The role is the unifying technical authority across hardware, software, compute, networking, and storage, and is responsible for architecting a cohesive, rack-scale AI platform optimized for trillion-parameter LLM training and high-throughput inference. The position spans advanced clustering and distributed training frameworks through physical infrastructure such as PCIe Gen 5/6 pathways, NVMe storage topologies, and RDMA fabrics, providing AI research and deployment teams with a powerful, integrated platform. The architect will define a long-term technical vision, guide platform and silicon requirements, analyze market competition and technology costs, and work across engineering disciplines to ensure software and hardware are designed and optimized together.

Requirements

  • At least 4 years functioning as a Lead or Principal Architect for large-scale AI or machine-learning platforms, with experience in systems engineering, cloud architecture, HPC, or hardware engineering.
  • Deep practical knowledge of training and deploying large models, including data, tensor, and pipeline parallelism and the infrastructure needs of modern LLM architectures.
  • Authoritative understanding of system-level bottlenecks and data pathways, including PCIe Gen 5/6, NVMe namespaces, and RDMA integration such as RoCEv2 or InfiniBand.
  • Experience with container orchestration platforms and infrastructure-as-code tailored to GPU-heavy bare-metal and cloud environments.
  • Ability to bridge AI researchers and data scientists with low-level hardware, CPU, memory, storage, GPU, and network engineers, translating model requirements into infrastructure specifications.
  • Ability to produce platform engineering requirement specifications that guide and influence future silicon designs.

Nice To Haves

  • Hands-on experience with rack-as-a-system AI platforms integrating current networking, cooling, and GPU technologies.
  • Working knowledge of scripting languages such as Python or JSON to characterize workloads on bare-metal AI compute systems and expose issues with neural-engine silicon solutions.

Responsibilities

  • Define holistic architecture for highly clustered AI environments, ensuring data flows effectively between parallel storage systems, AI compute nodes, and high-bandwidth network fabrics.
  • Influence AI workload scheduling and orchestration strategies using tools such as Kubernetes or Slurm for distributed training, model checkpointing, and inference serving.
  • Profile and remove system-level bottlenecks across the AI pipeline, including tuning deep learning frameworks, OS-level NUMA pinning, and I/O scheduling.
  • Collaborate with software, firmware, and OS engineering teams to ensure the software stack uses hardware capabilities effectively, including ARM mesh interconnects and merchant silicon features.
  • Drive the 3-to-5-year technical vision for the AI platform.
  • Collaborate with processor, memory, storage, GPU, thermal, mechanical, BIOS, and manageability specialists to define platform requirements and communicate them to internal and external silicon teams.
  • Influence silicon features, board-routing guidelines, power and thermal targets, and platform performance specifications.
  • Conduct AI industry and competitive analysis, including TCO analysis of OPEX and CAPEX for new technologies.

Benefits

  • Medical coverage
  • Dental coverage
  • Vision coverage
  • Flexible Spending Accounts (FSAs)
  • Health Savings Accounts (HSAs)
  • Disability insurance
  • Life insurance
  • 401(k) retirement plan
  • Commuter benefits
  • Wellness services
  • Employee Assistance Programme (EAP)
  • Flexible working
  • Inclusive work environment
  • Reasonable adjustments during interviews
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service