Full-Time

Staff SOC Performance Architect

Confirmed live in the last 24 hours

Tenstorrent

Tenstorrent

501-1,000 employees

Builds advanced computers for AI applications

Hardware
AI & Machine Learning

Compensation Overview

$100k - $500kAnnually

Senior, Expert

No H1B Sponsorship

Remote in USA

Role is remote, based out of the United States or the United Kingdom.

US Citizenship Required

Category
Embedded Systems Engineering
Electrical Engineering
Signal Processing Engineering
Required Skills
Python
CUDA
Pytorch
C/C++
Computer Vision
Requirements
  • Bachelor's or Master's degree in Electrical Engineering, Computer Engineering, or a related field. A Ph.D. is a plus.
  • Strong expertise in C/C++ and Python based modeling for SoCs
  • Solid understanding of at least 3 ML network architectures and their performance characteristics.
  • Experience with running ML models on GPU architectures.
  • Experience with CUDA / CUDNN / pytorch.
  • Solid understanding of computer architecture, interconnects and memory system behavior
  • Proven experience in modeling and analyzing complex workloads for SoCs.
  • Strong problem-solving skills and the ability to analyze complex system-level issues.
  • Excellent communication skills and the ability to collaborate effectively with cross-functional teams.
Responsibilities
  • Gather, refine and document Computer Vision, Machine Learning and other automotive workloads
  • Work with internal teams to map automotive workloads to Tenstorrent architecture.
  • Run automotive ML workloads on performance models. Analyze and summarize results.
  • Develop SoC and subsystem level performance use case descriptions.
  • Develop models of interconnect and memory subsystem behavior for target workloads
  • Analyze workload in detail and give guidance on improvements at both module and chip level.
  • Create clear documentation of findings and recommendations for performance improvement.
  • Present results to internal stakeholders and customers
  • Participate in competitive analysis, comparing achieved performance to other industry solutions

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 technologies like ASIC design and RISC-V architecture. Unlike many competitors, Tenstorrent emphasizes a combination of computer architecture expertise and neural network compilers to optimize AI performance. The company's goal is to advance the capabilities of AI computing, serving clients in the AI and computing sectors while generating revenue through the sale of their systems and services.

Company Stage

Late Stage VC

Total Funding

$1.3B

Headquarters

Toronto, Canada

Founded

2016

Growth & Insights
Headcount

6 month growth

15%

1 year growth

38%

2 year growth

131%
Simplify Jobs

Simplify's Take

What believers are saying

  • Tenstorrent raised $693M in Series D funding, enhancing its growth potential.
  • Partnership with BOS Semiconductors positions Tenstorrent in the growing automotive AI chip market.
  • RISC-V architecture's popularity aligns with Tenstorrent's technology focus, boosting its market relevance.

What critics are saying

  • Competition from NVIDIA could challenge Tenstorrent's growth in the AI data center market.
  • Reliance on partnerships like BOS Semiconductors may expose Tenstorrent to collaboration risks.
  • Ambitious expansion plans could strain resources and lead to operational challenges.

What makes Tenstorrent unique

  • Tenstorrent specializes in AI hardware and software for high-performance computing.
  • The company leverages RISC-V technology for flexible and open-source AI solutions.
  • 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?