System Software / Embedded Engineer

Tensordyne•Sunnyvale, CA

20h

About The Position

Tensordyne is an AI system solution company developing high-performance, low-power generative AI inference systems through custom silicon, hardware, and software. Its mission is to enable multimodal generative AI inference acceleration at scale for hyperscaler and neocloud data center customers. The company is a well-funded, fast-paced startup headquartered in Sunnyvale, California, and Munich, Germany, with team members working remotely across North America and Europe. The System Software / Embedded Engineer will join Tensordyne’s Platform Software team and lead requirements gathering, design, and development of system-level services, tools, and supporting tests for the AI inference platform software. The work includes debugging, profiling, and analysis capabilities, as well as supporting cross-functional teams during platform software bring-up and validation. The engineer will partner closely with hardware, silicon, and validation teams to deliver robust, production-quality platform software. The role works close to hardware on firmware and board bring-up, bootloaders, board support package (BSP) development, device drivers, and Linux kernel modules. It also includes hardware/software co-design and system-level performance and power optimization, using low-level interfaces such as PCIe, I2C, SPI, and JTAG. Occasional travel to partner, customer, or contract manufacturer sites may be required but is not mandatory.

Requirements

At least 5 years of experience developing low-level systems and embedded software, firmware, and tooling for high-performance computing systems deployed to data center customers.
Hands-on experience bringing up and debugging complex Linux-based systems at the kernel and driver level, including embedded Linux, device drivers, and/or RTOS environments.
Expert programming experience in C and C++.
Proficiency in Python, Bash/Shell, and/or related programming languages.
Comfort interfacing directly with hardware, reading schematics and datasheets, and debugging with tools such as JTAG, logic analyzers, and oscilloscopes.
Self-motivated and adaptable, with the ability to work across hardware, silicon, and software teams in a fast-paced startup environment.
Excellent technical communication skills, including the ability to explain complex technical concepts and practical solutions clearly.
BS or MS degree in Electrical Engineering, Computer Science, or a related technical field.

Nice To Haves

Experience with SoC/silicon bring-up.
Familiarity with AI inference frameworks and data center deployment/management environments.
Rust experience.

Responsibilities

Lead requirements gathering, design, and development of system-level services, tools, and supporting tests for Tensordyne’s AI inference platform software.
Develop debugging, profiling, and analysis capabilities.
Support cross-functional engineering teams in platform software bring-up and validation on AI inference systems.
Partner with hardware, silicon, and validation teams to deliver robust, production-quality platform software.
Work on firmware and board bring-up, bootloaders, BSP development, device drivers, and Linux kernel modules.
Collaborate with silicon and hardware teams on hardware/software co-design.
Optimize system-level performance and power across the AI inference platform.
Work with low-level interfaces including PCIe, I2C, SPI, and JTAG.