Staff CPU Microarchitect, Vector and Matrix Compute

About the role

The role defines next-generation RISC-V vector and matrix compute for HPC and AI workloads. It unites architecture and micro-architecture for execution units and subsystems. The position bridges hardware and software to shape how CPUs execute large-scale AI and HPC workloads. You will own the definition and execution of the CPU microarchitecture for matrix and vector compute, ensuring that the design meets demanding performance, power, and area targets for frontier workloads. You will collaborate closely with software teams to ensure that the architecture enables efficient programming models and that hardware features are validated through software development. This position requires you to make tradeoffs between complexity, performance, and risk while maintaining a clear roadmap for future generations of compute. You will be responsible for translating high-level product requirements into detailed microarchitectural specifications that can be implemented in advanced semiconductor processes.

Key facts

What you'll do

Define architecture and micro-architecture for RISC-V Vector extensions, floating point, and matrix compute units.

These designs serve future HPC and AI computing platforms and their evolving demands.

Cross-functional collaboration influences CPU cluster and subsystem decisions across the development lifecycle.

Leadership ensures alignment across hardware and software stakeholders.

The design emphasizes efficient data movement and system-level optimization.

Perform RTL design with cutting-edge AI frontier models as part of defining and validating compute architectures. RTL development supports implementation and verification of matrix and vector compute extensions.

Drive performance analysis and optimization of matrix multiplication, convolution, and reduction operations at the microarchitectural level.

Develop modeling frameworks and benchmarks to evaluate different design options and guide architecture decisions.

Requirements

Bring 5+ years of experience in CPU design for high-performance computing and AI acceleration. Experience must cover architecture and micro-architecture decisions for compute subsystems.

Demonstrate experience designing hardware for matrix floating-point computation, AI accelerators, vector processing, tensor engines, or machine learning workloads.

Show knowledge of CPU architecture, memory hierarchy, data movement, cache coherency, and interactions between CPUs and AI accelerators. Understanding must support system-level optimization for HPC and AI.

Prove experience defining architectural specifications and leading cross-functional teams from concept through silicon implementation. Track record includes managing requirements and design tradeoffs.

Have hands-on experience with RTL development targeting cutting-edge AI frontier models. Proficiency in hardware description and verification is required.

Practical notes

This role is remote and based in North America. Role is contingent on U.S. export-controlled technology eligibility and compliance with U.S. Typical interview steps

Leadership hiring usually includes a strategy case, a people scenario, and a values conversation. Candidates may be asked to turn around a struggling team or plan for growth. Past results are examined closely, especially how decisions were made. Interviewers dig into specific decisions you made and their consequences. Preparing two or three decision stories with numbers and lessons is the best use of time.

Good to know

Work focuses on open RISC-V computing, vector extensions, and matrix compute for AI and HPC. The role uses CPUs, AI accelerators, and memory subsystems in large-scale datacenter environments. Design outcomes influence data movement, system-level optimization, and hardware-software co-design.

Questions to ask

Useful questions for the interview: what a typical week looks like, how work is assigned, what tools the team uses, and how feedback works. Asking how the role has changed recently and what the team wishes it had known when joining is also reasonable. Questions about the manager's priorities are especially valued.

Career growth

Leadership careers move from manager to senior manager, director, and executive roles. Each step adds more scope and more responsibility for outcomes. The best leaders keep learning how to delegate, communicate, and decide. Leadership careers reward a record of building and developing people. The higher you go, the more your communication shapes the whole organization.