Simplify Logo

Full-Time

Performance Architect

Confirmed live in the last 24 hours

Tenstorrent

Tenstorrent

201-500 employees

Develops AI/ML hardware accelerators and software

Hardware
AI & Machine Learning

Compensation Overview

$100k - $500kAnnually

Mid, Senior

Remote in USA

US Citizenship Required

Category
Hardware Engineering
Computer Hardware Engineering
Hardware Validation & Testing
Required Skills
Verilog
Python
Git
VHDL
Requirements
  • BS/MS/PhD in EE/ECE/CE/CS
  • Demonstrated experience in CPU, SOC, or ASIC architectural performance modeling and simulation.
  • Hands-on experience with C or System C performance modeling.
  • Hands-on experience with implementing synthesizable traffic generators.
  • Hands-on experience with collecting workload traces and mapping those to traffic generator configurations.
  • Strong understanding of traffic patterns, timings, priority classes, and dependencies.
  • Strong understanding of performance bottleneck sources and mitigation techniques.
  • Proficiency in hardware description languages (HDL) such as Verilog or VHDL.
  • Proficiency in programming languages and tools including Python and Git.
  • Strong analytical and problem-solving skills.
  • Excellent communication and collaboration skills.
Responsibilities
  • Develop a scalable architectural performance modeling strategy for use across a wide range of product lines.
  • Steer the architectural choices towards the right solution to meet the performance targets.
  • Collect workload traces.
  • Identify under/over utilization of hardware resources.
  • Identify bottlenecks in the fabric and memory hierarchy and quantify the performance gains from specific changes to the architecture or implementation.
  • Correlate performance modeling results to emulation results.

Tenstorrent specializes in developing high-quality AI/ML accelerators, including individual PCIe cards, workstations, servers, and ultra-dense Galaxy pods, featuring industry-standard chiplets and a modular RISC-V CPU. Their products also incorporate a BUDA software framework designed for open-source collaboration, enabling efficient and scalable hardware for deep learning applications.

Company Stage

Series C

Total Funding

$334.5M

Headquarters

Toronto, Canada

Founded

2016

Growth & Insights
Headcount

6 month growth

13%

1 year growth

30%

2 year growth

116%
Simplify Jobs

Simplify's Take

What believers are saying

  • The launch of next-generation Wormhole-based developer kits and workstations could attract a significant developer community, driving innovation and adoption.
  • Collaborations with industry giants like Hyundai and Rapidus indicate strong growth potential and access to advanced manufacturing technologies.
  • The introduction of specialized AI inference acceleration boards like the Grayskull e75 and e150 can capture a niche market in AI and machine learning applications.

What critics are saying

  • The competitive landscape in AI hardware is intense, with major players like NVIDIA and Intel posing significant challenges.
  • Dependence on strategic partnerships for advanced manufacturing and technology development could lead to vulnerabilities if these partnerships falter.

What makes Tenstorrent unique

  • Tenstorrent's use of RISC-V architecture in their AI processors offers a unique alternative to traditional x86 and ARM architectures, providing flexibility and open-source benefits.
  • Their focus on high-performance AI chips and scalable developer kits positions them as a key player in the AI hardware market, particularly for developers seeking robust multi-chip solutions.
  • Strategic partnerships with global entities like Rapidus and C-DAC enhance their capabilities in cutting-edge semiconductor technology and edge AI processing.