SoC Performance Architect

Charterhouse is working with an innovative technology startup specializing in the design and development of advanced, customized server hardware solutions. The company has a big focus on ML (Machine Learning) applications and AI (Artificial Intelligence). They are growing and looking to hire an SoC Performance Architect.

This role as a Performance Architect involves defining the architecture for the next generation of Machine Learning ASICs, developed using advanced process technologies and designed to offer leading performance and feature sets.

The role will have various responsibilities, including working on performance analysis, modelling and architecture definition of ML ASICs. This role also sits within a cross-function team working closely with designers and software engineers with a focus on program management.

The ideal candidate will have more than 5 years of experience in defining and delivering high-performance ASICs into production, with a particular emphasis on architecture definition and performance analysis. Proficiency in programming with C / C++ or SystemC and the ability to map hardware algorithms to efficient code are essential. The successful candidate will need to have a strong understanding of computer architecture concepts, such as microprocessor architecture, memory systems, on-chip interconnection networks, and hardware / software partitioning.

Previous exposure to designing HP AI accelerators like GPUs, NPUs, or TPUs is highly beneficial. Additionally, experience with deep neural network (DNN) architectures, workloads, and frameworks is a bonus, as is knowledge of low-power design techniques and advanced process nodes. Excellent communication skills in English are essential for this position.