Senior Software Engineer, Compute Platform Integration

Job description

Job responsibilities

• Own the boot path end to end on every compute generation we ship: firmware, UEFI, the unified kernel image and its command line, initrd, encrypted root unlock, and the handoff into a fully running NixOS system.
• Guarantee full hardware availability at boot: every bus, peripheral, and network interface present, correctly named, and functional, with deterministic enumeration that survives platform and kernel upgrades.
• Own carrier board device trees: reading pinout and pinmux definitions from hardware engineering, producing and maintaining the corresponding DTS, DTSI, and overlay changes, and validating them on real boards.
• Produce kernel, bootloader, and BSP patches when our hardware needs behavior the vendor tree doesn’t provide, and carry those patch sets forward across platform upgrades.
• Serve as our primary technical liaison to the platform organization: file and drive issues with clear reproductions, review upstream changes for impact on our hardware, and land product-side integration for new…
• Sustain deployed compute generations through vendor lifecycle transitions, applying security patches and backports so fielded systems stay stable and supportable while newer platforms come online.
• Support new hardware bring-up alongside electrical engineers and our platform team, owning the boot and device tree side of getting a new board to a validated, production-flashable state.
• Debug the hard ones: intermittent bus errors, thermal and power-related faults, boot hangs, and failures that only reproduce on one unit in the field.

Job requirements

• 4+ years of professional software engineering with substantial embedded Linux platform, board support, or kernel-adjacent integration work.
• Deep understanding of the embedded Linux boot chain: firmware and bootloader stages, UEFI, kernel and initrd, kernel command line, root filesystem mounting, and how each stage fails.
• Device tree fluency: reading and modifying DTS, DTSI, and overlays for custom hardware, and connecting a pinout or pinmux specification to the resulting device tree.
• Ability to read, patch, and debug kernel and BSP source in C, plus proficiency in Rust or C++ for the userspace services that depend on your platform.
• Practical peripheral and bus knowledge across CAN, I2C, SPI, UART, PCIe, USB, and multi-port Ethernet, sufficient to determine whether a device is missing, misconfigured, or broken.
• Comfort debugging real hardware: serial consoles, boot logs, dmesg archaeology, and reading schematics and datasheets to answer your own questions.
• Demonstrated success driving technical work through another team, with the clarity and persistence to get a platform or vendor issue resolved rather than worked around.
• Comfort owning software on hardware that is already deployed, where a regression reaches real systems and a fix has to be safe as well as correct.