Interconnect and Compute Architect
About the role
Tenstorrent is seeking a Interconnect and Compute Architect to help define and build next-generation CPU networking architecture for both datacenter and emerging robotics/automotive applications. The initial focus will be datacenter networking, with robotics as the first target within the automotive/robotics space.
Responsibilities
- Contribute to our current datacenter networking efforts while also helping to seed and specify future medium- to low-power robotics/automotive devices for AI/ML compute and sensor ingest.
- Build next-generation CPU networking architectures for both high-performance datacenters.
- Collaborate across hardware, software, and systems teams to define and refine networking requirements.
- Guide architectural decisions through the intersection of forwarding, buffering, modeling, and RTL design.
- Understand and implement AI/ML specific protocols and implementations.
- Learn how to build extreme low latency networks across ethernet topologies and high performance computing (HPC) topologies.
Requirements
- Knowledge of Ethernet network architecture and how performance is modeled.
- Experience with die-to-die interfaces and understanding associated protocols and design tradeoffs.
- Understanding of Ethernet networking concepts and their mapping onto on-chip and off-chip fabrics.
- Experience with datacenter scale up architectures like UALink, NVLink, and Broadcom SUE.
- Experience with scale out RDMA protocols like RoCE, Infiniband, and others.
Qualifications
- You thrive while navigating multiple priorities and ambiguous, evolving requirements.
- You have experience with die-to-die interfaces and understanding associated protocols and design tradeoffs.
- You understand Ethernet networking concepts and their mapping onto on-chip and off-chip fabrics.
- You have experience with datacenter scale up architectures like UALink, NVLink, and Broadcom SUE.
- You have experience with scale out RDMA protocols like RoCE, Infiniband, and others.
Skills
- Design and implementation of next-generation CPU networking architectures.
- Collaboration across hardware, software, and systems teams.
- Guide architectural decisions through the intersection of forwarding, buffering, modeling, and RTL design.
- Implementing AI/ML specific protocols and implementations.
- Building extreme low latency networks across ethernet topologies and HPC topologies.
Benefits
Compensation for all engineers at Tenstorrent ranges from $100k - $500k including base and variable compensation targets. Experience, skills, education, background and location all impact the actual offer made.
Pay
Compensation for all engineers at Tenstorrent ranges from $100k - $500k including base and variable compensation targets. Experience, skills, education, background and location all impact the actual offer made.
Schedule
This role is remote, based out of North America.
Qualifications
- Candidates must be eligible to access U.S. export-controlled technology.
- The offer may be contingent upon your citizenship/permanent residency status or ability to obtain prior license approval from the U.S. Commerce Department or applicable federal agency.
Equal Opportunity Employer
We are an equal opportunity employer.