Latest AI JobsBrowse Jobs
Tenstorrent logo

Principal CPU Microarchitect - RISC-V & AI Compute

TenstorrentAustin, Texas, United States; Santa Clara, California, United States; Toronto, Ontario, CanadaAI Hardware: Tensix
Apply on company site
Salary
Not listed
Job type
Full-time
Posted
28 Jul 2025
Status
Open · checked 7d ago

About this role

Tenstorrent is seeking a Principal CPU Microarchitect to join their hybrid AI Hardware Tensix team in Austin, Texas, Santa Clara, California, or Toronto, Ontario. This role focuses on shaping the core CPU architecture that powers advanced artificial intelligence platforms and suits experienced hardware professionals passionate about RISC-V technology.

Role Overview

Tenstorrent is pioneering advanced artificial intelligence technology by combining software models, compilers, platforms, networking, and custom semiconductors. The Tensix team is building the future of AI compute using a scalable RISC-V processor architecture designed entirely from the ground up. Professionals in this hybrid position will work directly on reimagining the frontend of RISC-V cores to maximize developer experience, efficiency, and programmability. Candidates at various seniority levels are welcome to apply, with final levels determined during the interview process.

Key Responsibilities

Dissect state-of-the-art AI workloads to uncover execution bottlenecks and define exact CPU performance requirements.
Lead the microarchitecture design of custom RISC-V CPU cores while balancing strict Performance, Power, and Area (PPA) goals.
Create and implement custom instruction set architecture (ISA) extensions that deliver substantial performance gains without hurting compiler usability or developer velocity.
Collaborate closely with software, compiler, and hardware engineering departments to co-design next-generation artificial intelligence compute systems.

Requirements

Bring ten or more years of deep professional expertise in CPU microarchitecture design and performance modeling.
Possess thorough familiarity with the computational limits and memory bottlenecks associated with modern AI workloads, specifically Large Language Models (LLMs).
Demonstrate a strong drive for hardware-software co-design by architecting custom instruction set extensions and validating performance improvements against real-world tasks.
Meet eligibility requirements for accessing U.S. export-controlled technology, which may involve citizenship or permanent residency status or the ability to obtain necessary federal export licenses.

Nice to Have

Practical familiarity with open-source RISC-V core designs.
Prior experience working with AI-based agentic workflows.

Skills and Tools

RISC-V CPU microarchitecture design
CPU performance modeling
Instruction Set Architecture (ISA) development
AI workload profiling and bottleneck analysis
Hardware-software co-design methodologies

Salary and Benefits

Tenstorrent provides a highly competitive compensation package along with comprehensive benefits for eligible employees.

About the Company

Tenstorrent is an industry leader in cutting-edge artificial intelligence technology, focused on revolutionizing cost efficiency, ease of use, and raw performance expectations. The company employs a diverse team of technologists dedicated to building the best possible AI platform through curiosity, collaboration, and a commitment to difficult problem-solving.

How to Apply

Interested candidates should submit their applications directly on the official Tenstorrent careers page.

Questions candidates ask

What locations are available for this position?

This hybrid role is based out of Austin, Texas, Santa Clara, California, or Toronto, Ontario.

How many years of experience are required?

Applicants need 10+ years of deep expertise in CPU performance modeling and microarchitecture design.

Are candidates with different seniority levels considered?

Yes, Tenstorrent welcomes candidates at various experience levels, assessing them during the interview process to align offers appropriately.

What specific AI workloads are a focus of this role?

The position heavily involves working with modern AI workloads, with a particular focus on Large Language Models (LLMs).

Are there export control requirements for the job?

Yes, employment is contingent upon the applicant's eligibility to access U.S. export-controlled technology due to compliance with U.S. export laws.

Ready to apply?

The application is on the official Tenstorrent careers page.

Apply on company site