Architect

Job Title

Architect

Role Summary

Design and deliver performant silicon architecture at the full-chip or subsystem level for high-performance and GenAI workloads. Work across research, software, and hardware teams to translate workload requirements into architecture specifications and implementation trades.

Experience Level

Level - Senior. The posting does not state explicit years of experience but expects significant experience in high-performance CPU, GPU, or AI accelerator architecture and hardware/software interactions. This is a hybrid role requiring in-office presence Tuesday–Thursday in Mountain View, CA.

Responsibilities

Accountable for architecture definition, trade-offs, and cross-team coordination throughout the product lifecycle.

• Develop, support, and own hardware architecture across the product lifecycle.
• Capture customer use cases and workload performance goals; translate into architecture requirements and specifications.
• Estimate hardware area, timing, and power to inform architectural trade-offs.
• Collaborate closely with Research, Software, and Hardware teams and interface with vendors and customers.
• Design for scalable data movement, memory and network fabrics, and parallel compute models.
• Analyze workload mapping, identify bottlenecks, and propose system-level solutions.

Requirements

Key must-have qualifications and useful additional skills.

• Must-have: Proven expertise in modern high-performance CPU, GPU, or AI accelerator architecture and hardware/software interactions.
• Must-have: Experience analyzing and optimizing system performance using performance tools and methodologies.
• Must-have: Proficiency in programming and code optimization for prototyping and performance analysis.
• Must-have: Ability to produce clear requirements and architecture specifications; estimate area, timing, and power.
• Must-have: Strong written communication skills and experience collaborating with cross-functional teams.
• Must-have: Authorized to work in the United States and able to work from the Mountain View office Tuesday–Thursday; must be able to comply with U.S. export control requirements without obtaining a license.
• Nice-to-have: Experience with SIMD, Vector, VLIW, or other parallel execution models.
• Nice-to-have: Experience designing efficient data movement across memory and network fabrics, and with distributed systems and workload mapping for scalable systems.
• Nice-to-have: Background in ML systems research or experience working directly with ML workloads.

Education Requirements

Not specified.

About the Company

Company: MatX

Headquarters: Mountain View, California, USA

MatX specializes in creating faster chips for large language models (LLMs), focusing on innovative hardware and software solutions. The company fosters a collaborative and supportive work environment, welcoming candidates of all experience levels. Their approach prioritizes deep understanding and consideration of novel methods to drive efficiency and performance in their projects, particularly in silicon design and related engineering roles.