AI System Architect

AI System Architect Overview

An exciting opportunity has arisen for an AI System Architect to join a pioneering startup in Oxfordshire, England, that is transforming the landscape of computing through the development of an AI accelerator. This innovative company is focused on creating technology that enhances the speed and efficiency of data center computations while significantly reducing energy consumption.

Key Responsibilities:

• Design and architect systems or chips specifically for machine learning training and inference.
• Conduct performance modeling including roofline analysis, bandwidth modeling, and cycle-level estimation.
• Demonstrate fluency in hardware-software integration, including instruction set architecture (ISA) design, compiler interfaces, runtimes, and datapath microarchitecture.
• Influence product direction through the creation of strong written specifications and data-driven decision-making.

Requirements:

• Extensive expertise in artificial intelligence and machine learning, particularly with transformers, convolutional neural networks (CNNs), and recommender systems.
• Proven experience in the AI hardware industry, particularly with AI-chip startups or major AI-hardware teams such as TPU, MTIA, Trainium, or Dojo.
• Familiarity with compiler stacks including MLIR, XLA, TVM, and Triton, and understanding how these influence hardware design choices.

Benefits:

• Competitive salary and equity options.
• Flexible working arrangements to support work-life balance.
• Opportunity to work on cutting-edge problems in AI-driven silicon design.
• Immediate impact on product direction from the start.

If you are interested in this role and meet the qualifications, please reach out to Chris Amison for more information.

Degree Requirement: Not Specified

Visa Sponsorship May be