Full-Time

CPU Architect

Posted on 1/15/2025

Tenstorrent

Tenstorrent

501-1,000 employees

Builds advanced computers for AI applications

Hardware
AI & Machine Learning

Compensation Overview

$100k - $500kAnnually

Senior, Expert

United States

Role can be remote or in-office in the United States.

US Citizenship Required

Category
Hardware Engineering
Computer Hardware Engineering
Required Skills
Python
Assembly
C/C++
FPGA
Requirements
  • 7+ years of experience in silicon performance validation or CPU architecture.
  • Master’s degree in Computer Engineering, Computer Science, or a related field (or equivalent experience).
  • Strong expertise in CPU performance analysis, silicon validation, and computer architecture, with at least 5 years focused on performance validation.
  • Proficiency in pre-silicon performance models, RTL, emulation, and FPGA platforms (e.g., Zebu, HAPS).
  • Hands-on experience with tools like Linux perf, ARM DS5, or custom validation tools.
  • Skilled in writing and analyzing microbenchmarks using assembly and/or C/C++.
  • Proficiency in C++, Python, and SystemVerilog for scripting and validation purposes.
  • Familiarity with Continuous Integration (CI) systems and task automation using Python.
Responsibilities
  • Write vector microbenchmarks to create power viruses and test vector performance on current and future CPU microarchitectures.
  • Work with SW & compiler team on analyzing and optimizing vector compilation and performance on multiple important workloads.
  • Participate in the RISCV International workgroups related to vector specifications as a representative of Tenstorrent to give feedback on new vector instructions.
  • Explore hardware and software co-optimization opportunities to drive system performance improvements.
  • Propose PPA optimizations for the current vector microarchitecture, including changes to load/store microarchitecture and instruction cracking.
  • Define and enhance methodologies for pre-silicon and post-silicon validation of high-complexity System-on-Chip (SoC) designs.
  • Innovate techniques for performance validation, ensuring that products meet and exceed specification requirements.
  • Work closely with architects, hardware engineers, and microcode (ucode) engineers to understand and implement new CPU features.
  • Analyze SoC design features to assess their performance impact and identify optimization opportunities.
  • Investigate key workloads, pinpoint bottlenecks, and prototype solutions to improve performance.
  • Design and maintain the silicon validation architecture roadmap for next-generation product introductions.
  • Define validation requirements in collaboration with cross-functional teams, ensuring compliance throughout the design lifecycle.
  • Optimize silicon validation architectures for performance, power consumption, and area efficiency.
  • Develop and maintain complex models and simulations of silicon devices using advanced tools.
Desired Qualifications
  • Demonstrated knowledge of Vector architecture is highly desirable.

Tenstorrent builds advanced computers specifically designed for artificial intelligence applications. Their products include high-performance computing systems that utilize specialized hardware and software solutions, focusing on areas like computer architecture, ASIC design, and RISC-V technology. These systems are tailored to meet the needs of clients in the AI and computing sectors. What sets Tenstorrent apart from its competitors is its deep expertise in neural network compilers and its commitment to developing systems that enhance AI performance. The company's goal is to lead advancements in AI computing by providing powerful and efficient computing solutions.

Company Stage

Series D

Total Funding

$1.3B

Headquarters

Toronto, Canada

Founded

2016

Growth & Insights
Headcount

6 month growth

-3%

1 year growth

-6%

2 year growth

-4%
Simplify Jobs

Simplify's Take

What believers are saying

  • RISC-V architecture's growth aligns with Tenstorrent's open-source AI software focus.
  • Automotive AI chip market growth offers lucrative opportunities for Tenstorrent's BOS partnership.
  • Increasing demand for HPC systems could expand Tenstorrent's reach beyond tech industries.

What critics are saying

  • Competition from established AI hardware companies like Nvidia could impact market share.
  • Reliance on open-source software may delay product development timelines.
  • Geopolitical tensions in South Korea and Japan could disrupt Tenstorrent's supply chains.

What makes Tenstorrent unique

  • Tenstorrent leverages RISC-V technology for flexible, open-source AI hardware solutions.
  • The company partners with BOS Semiconductors to develop next-gen automotive AI chips.
  • Tenstorrent's global presence includes offices in key tech hubs like Silicon Valley and Tokyo.

Help us improve and share your feedback! Did you find this helpful?

Benefits

Hybrid Work Options