Senior Embedded Software Engineer

The role

We are looking for an Embedded Software Developer to design and implement the firmware and low-level software that powers our next-generation GPU and HPC platforms. This role will focus on embedded control, board management, telemetry, and hardware-firmware integration, ensuring that our systems operate reliably in high-density, mission-critical environments.

Key Responsibilities

• Design and implement embedded firmware for server management, telemetry, and control systems.

• Maintain and enhance our custom OpenBMC firmware with new features and improvements

• Enable real-time monitoring of power, thermal sensors, and hardware health.

• Work closely with hardware engineers to validate firmware for existing and future platforms

• Debug and optimize low-level drivers and protocols.

• Contribute to long-term firmware architecture for GPU cluster reliability.

Required Skills & Qualifications:

• 5+ years in embedded systems or firmware development.

• Proficiency in embedded Linux.

• Hands-on experience with BMCs, microcontrollers, or SoC firmware.

• Understanding of hardware bring-up and debugging

• Languages: C, C++, Bash, Go, YAML

• Firmware: OpenBMC, U-Boot, Linux Kernel

• Interfaces: I2C, I3C, SPI, eSPI, UART, LPC

• Protocols: SMBus, PCIe, PMBus, PECI

• Build Systems: Meson, CMake

• Descriptors & Formats: FRU, SMBIOS, ACPI, DMI

Preferred:

• Knowledge of the Yocto Project principles

• Knowledge of systems and D-Bus principles

• Proficiency in C++

• Good knowledge of C, sufficient for periodic work with Linux drivers and the U-Boot bootloader;

• Experience in developing Linux drivers of any kind, especially those implementing sysfs and hwmon interfaces.

• Experience with server BMC firmware IPMI, IPMB, KCS, SSIF, Redfish, PLDM

• Knowledge of GPU/CPU telemetry frameworks (e.g., NVML, DCGM)

• Exposure to firmware security (Secure Boot, signed firmware)

• Experience with RAS (Reliability, Availability, Serviceability)

• Background in high-performance computing or data center hardware.