AI Modelling Architect

Location: San Jose
Industry: Semiconductors, AI Computing & Advanced Hardware
Salary: Competitive, dependent on experience

We are working with an innovative technology organisation developing advanced computing solutions for artificial intelligence and machine learning applications.

They are looking to hire an experienced AI Computing Architect to contribute to the architecture and development of next-generation semiconductor technologies. This is a hands-on technical role focused on defining system-on-chip (SoC) and intellectual property (IP) architectures for specialised AI computing hardware.

The successful candidate will work at the intersection of hardware architecture, machine learning and software engineering, helping to develop and optimise hardware acceleration solutions for AI training and inference.

Working closely with cross-functional engineering teams, you will contribute to architectural decisions, performance modelling and hardware-software optimisation, ensuring that future computing platforms meet demanding performance, power and efficiency requirements.

Essential:

• PhD in Electrical Engineering, Computer Science or a related discipline, with at least three years of relevant experience; alternatively, a Master's degree with five or more years of relevant experience.
• Strong programming skills in C/C++ and Python.
• Solid understanding of modern system-on-chip (SoC) architectures and microarchitecture principles.
• Knowledge of performance, power and area (PPA) trade-offs in digital hardware design.
• Understanding of fundamental deep-learning algorithms, including convolutional neural networks (CNNs) and recurrent neural networks (RNNs).
• Experience implementing deep-learning algorithms in hardware.
• Ability to analyse computing requirements and translate application needs into hardware architecture solutions.

Required / Highly Preferred:

• Experience with SoC or IP architecture for AI computing applications.
• Knowledge of hardware and software co-design, including interactions between AI workloads, computing architectures and software frameworks.
• Experience with architectural performance modelling, analysis and optimisation.
• Ability to collaborate effectively with ASIC design, validation and software engineering teams.

Beneficial:

• Experience in embedded programming and firmware development.
• Familiarity with CUDA kernel development and GPU programming.
• Experience developing functional models to support hardware and compiler validation.

This is an opportunity to work on technically challenging AI computing technologies, contributing to architectural decisions that shape the development of advanced hardware platforms. You will collaborate with specialists across hardware and software disciplines, helping bridge the gap between emerging machine-learning requirements and efficient semiconductor implementations.