You and the process

With a person
Anna, in-house recruiter
9 years hiring engineers
30 minutes with a real recruiter
They read your CV with you, on a call, and say where the offers are being lost.
Didn’t find what you were looking for? Tell us what to build
Graphcore

AI Platform Architect

  • Office
  • 3-6 years

Salary

Not stated

Similar roles pay $100K - 170K a year · our estimate

AI summary

For members

The whole posting in a few lines. Sign up to read it here and on every role you open.

Sign Up to Read

Description

The company develops hardware, software and systems infrastructure for AI compute, aiming at the next wave of AI breakthroughs and wider adoption across industries.

It is part of a parent group whose stated goal is to enable artificial super intelligence and make its benefits available to everyone.

Teams combine AI researchers, silicon designers, software engineers and systems architects from varied backgrounds, and work in a culture of ongoing learning and innovation.

This role is the single technical lead across hardware, software, compute, network and storage for a rack-scale AI platform built for trillion-parameter LLM training and high-throughput inference.

The aim is a smooth, fast platform for AI research and deployment teams, covering everything from clustering and distributed training down to the physical layer, including PCIe Gen 5/6, NVMe storage layouts and RDMA fabrics.

Responsibilities

  • Define the overall architecture for tightly clustered AI environments so data moves between parallel storage, compute nodes and high-bandwidth networks without bottlenecks.
  • Shape the scheduling and orchestration strategy using Kubernetes or Slurm for distributed training, checkpointing and inference serving at very large scale.
  • Profile the full AI pipeline and remove system bottlenecks, from PyTorch and DeepSpeed down to NUMA pinning and I/O scheduling.
  • Work with software, firmware and OS engineers so the stack fully uses the hardware, including ARM mesh interconnects (RNI, HNF, SNF) and merchant silicon features.
  • Set the technical direction for the platform over the next 3 to 5 years. Work with processor, memory, storage, GPU, thermal, mechanical, BIOS and manageability experts to write requirement specifications, present them to internal and external silicon teams, and press for features, board routing guidelines, power and thermal targets and the right performance figures.
  • Run competitive market analysis and total cost of ownership (OPEX and CAPEX) studies of new technologies, which requires deep knowledge of the AI industry.

Requirements

  • Essential: 4+ years as a Lead or Principal Architect on large-scale AI or machine learning platforms, with a background in systems engineering, cloud architecture, HPC or hardware engineering.
  • Essential: Practical understanding of how large models are trained and deployed, including data, tensor and pipeline parallelism and the infrastructure modern LLM architectures need.
  • Essential: Thorough grasp of system bottlenecks and data paths, including PCIe Gen 5/6, NVMe namespaces and RDMA integration with RoCEv2 or InfiniBand.
  • Essential: Experience with container orchestration and infrastructure as code for GPU-heavy bare-metal and cloud environments.
  • Essential: Ability to bridge AI researchers and data scientists with CPU, memory, storage, GPU and network engineers, turning model needs into strict infrastructure specifications and platform requirement specifications that guide future silicon.
  • Desirable: Hands-on experience with rack-as-a-system GPU AI platforms.
  • Desirable: Experience with AI platforms that use the latest networking, cooling and GPU technologies on the market.
  • Desirable: Working knowledge of scripting in Python or JSON to characterize workloads on bare-metal AI compute systems and expose issues in current neural engine silicon.

Conditions

Flexible working is offered.

The company offers an equal opportunity process, and interviews can be adapted on request.

Benefits

  • A competitive salary.
  • Medical, dental and vision coverage.
  • Flexible Spending Accounts (FSAs) and Health Savings Accounts (HSAs).
  • Disability and life insurance.
  • A 401(k) retirement plan.
  • Commuter benefits.
  • Wellness services.
  • Employee Assistance Program (EAP) access.

Where you’d work

From the office

About the company

Graphcore

Office in Austin, United States

Also hiring in Bristol, United Kingdom, United States, Gdansk, Poland and 2 more places

Your chances

We've checked whether it's still hiring and how crowded it is.

  • 19 checks run
  • 0 red flags

Still hiring?

13 checks

No red flags

How crowded?

6 checks

No red flags

Fits Me

How well does this role fit you?

Answer a few questions or drop your CV, and every role gets a fit score with the reasons, this one first.

  • Your field
  • Level
  • Stack
  • Work model
  • Salary floor
  • Must-haves
Details8 facts · Role, Location, Compensation
Tech stack
  • Python
  • Kubernetes
  • PyTorch
  • LLMs
Region
United States
Pay period
Annual
Show 5 more factsShow less

Role

Category
Development
Experience
4+ years
Tech stack
  • Python
  • Kubernetes
  • PyTorch
  • LLMs

Location

Work model
Office
Region
United States
Office
  • Austin, United States

Compensation

Salary
Salary by agreement
Pay period
Annual

Something wrong with this vacancy?

Similar vacancies

Share this vacancy

What's wrong with it?

The employer never sees who reported.

Reason