Senior Engineer, Systems Debug (AI)

Location: Hybrid, based out of Toronto, Canada

Company

Tenstorrent develops high-performance AI accelerator platforms, RISC-V CPUs, software, networking, and semiconductor technologies.

What you will do

• Diagnose, root-cause, and resolve hardware and firmware issues across AI accelerator platforms, including chips, PCIe cards, and Galaxy systems.
• Lead silicon, board, and system-level debug efforts across power delivery, thermal behavior, GDDR memory, Ethernet/SerDes, PCIe, and NoC interconnects.
• Develop Python and C debug scripts, diagnostic tools, and automation frameworks for failure analysis and triage.
• Use oscilloscopes, logic analyzers, protocol analyzers, and other laboratory equipment to investigate system behavior.
• Interpret schematics, datasheets, register states, telemetry data, and firmware behavior to identify complex issues.
• Collaborate with ASIC design, firmware, software, and validation teams to improve product reliability and performance.

Requirements

• Experience in post-silicon validation, silicon bring-up, and system-level hardware debugging.
• Strong programming skills in Python and C.
• Hands-on laboratory experience with hardware debug equipment.
• Ability to analyze failures across silicon, boards, firmware, software, and high-speed interfaces.
• Ability to interpret schematics, datasheets, register states, telemetry data, and firmware behavior.
• Ability to work in a hybrid setup based in Toronto, Canada.

Culture & Benefits

• Collaborative environment focused on curiosity, difficult technical problems, and building reliable AI computing platforms.
• Exposure to AI accelerator architectures from prototype silicon through production.
• Opportunity to work across silicon, board design, firmware, drivers, and system software.
• Learning opportunities involving GDDR/DDR memory, PCIe, Ethernet/SerDes, power management, and system validation.
• Competitive compensation package and benefits.
• Employment is contingent on eligibility to access U.S. export-controlled technology; citizenship, permanent residency, or prior license approval may be required.