Performance Architect, AI HW at Tenstorrent. About the role Tenstorrent is building high-performance compute fabrics for AI and machine learning. This role focuses on modeling and optimizing hardware architectures to ensure maximum efficiency for complex AI workloads. Key facts
Location Toronto, Ontario, Canada; Austin, Texas; or remote
Engagement Full-time
Team Tensix
What you'll do Evaluate AI workloads across single and multi-node hardware setups to influence future architectural designs.
Build and support performance simulators, models, and micro-benchmark suites to test new features.
Execute PPA studies to weigh design tradeoffs and support hardware-software co-design.
Work with RTL, compiler, and runtime groups to align performance models with actual silicon outcomes.
Requirements Proficiency in C++ and Python for performance analysis, simulation, and modeling in heterogeneous systems.
Ability to identify system-level bottlenecks and analyze AI workload behavior.




Experience communicating technical trade-offs between hardware and software teams.
Analytical mindset focused on data-driven design for scalable AI systems.
Skills & tools C++
Python
Performance modeling
PPA analysis
AI workload benchmarking
Hardware-software co-design
Practical notes Compensation ranges from $100,000 to $500,000 including base and variable pay, determined by experience, skills, education, and location. Employment is contingent on eligibility to access U.S. export-controlled technology. Candidates may need to meet requirements related to citizenship or permanent residency status to comply with U.S. Export Administration Regulations, including potential licensing from the U.S. Commerce Department.
Tenstorrent is an equal prospect employer.

#J-18808-Ljbffr

📌 Performance Architect, AI HW (Ontario)
🏢 Tenstorrent
📍 Ontario