About Eridu AI 

Eridu AI is a Silicon Valley-based hardware startup pioneering infrastructure solutions that accelerate training and inference for large-scale AI models. Today’s AI performance is frequently limited by system-level bottlenecks. Eridu AI delivers multiple industry-first innovations across semiconductors, software, and systems to unlock greater GPU utilization, reduce capital and power costs, and maximize data center efficiency. The company’s solutions and value proposition have been validated by several leading hyperscalers.

The company is led by a veteran team of Silicon Valley executives and engineers with decades of experience in state-of-the-art semiconductors, optics, software, and systems, including serial entrepreneur Drew Perkins, co-founder of Infinera (NASDAQ: INFN), Lightera (acq. by Ciena), Gainspeed (acq. by Nokia) and Mojo Vision (World’s leading micro-LED company and developer of the first augmented reality contact lens).

Position Overview

We are seeking a highly skilled BSP Platform Engineer to lead the design and development of the Linux-based Board Support Package (BSP) within our platform software organization. In this role, you will architect, implement, and optimize the BSP stack for high-scale networking platforms, ensuring robustness, reliability, and performance across diverse hardware configurations.

Key Responsibilities

• Design, implement, and maintain high-quality Linux BSP and low-level platform software, including bootloaders, kernel subsystems, device drivers, and essential system services.
• Lead secure boot architecture and implementation, including chain-of-trust design, key provisioning, firmware authentication, measured boot, and integration with TPM/Root-of-Trust hardware.
• Drive early hardware bring-up and board initialization for new platforms, enabling stable operation of peripherals such as EEPROM, CPLD, GPIO, I2C, SPI, UART, PCIe, thermal sensors, and watchdogs.
• Validate functionality and performance across multiple hardware spins, including diagnostics, power-on self-tests, interface validation, and regression testing on supervisor/management boards.
• Collaborate closely with hardware, system, security, and manufacturing teams to define and optimize platform behavior around thermal management, power sequencing, reliability, and system resilience at the BSP layer.
• Enable control-plane software integration with SONiC platform services or custom management APIs, ensuring robust communication between the management CPU/BMC and the switching ASIC.
• Support production readiness through debugging, root cause analysis, DVT/EVT support, and long-term maintainability improvements

Required Qualifications

• Bachelor’s in Computer Science/ECE with 10+ years of experience, or Master’s with 5+ years in embedded systems, firmware, or platform software development.
• Deep proficiency in C/C++ programming in Linux-based cross-compiled environments, including familiarity with compiler toolchains (Yocto/Buildroot), debugging tools (gdb, JTAG, Lauterbach), and performance profiling.
• Strong expertise in embedded Linux, BSP development, board initialization, FPGA-to-CPU/BMC communication protocols and low-level system software, including bootloaders, kernel bring-up, device drivers.
• Hands-on experience with secure boot, chain of trust, key provisioning, TPM/Root-of-Trust integration, verified boot, and firmware authentication.
• Experience with hardware bring-up on new boards and SoCs, including DDR initialization, power sequencing, clock/reset trees, and low-speed peripheral drivers (I2C, SPI, UART, GPIO, EEPROM, CPLD).
• Experience developing platform SDK components, including board configuration frameworks, diagnostics, platform abstraction layers, and device-tree-based initialization.
• Experience with control-plane integration for network switches/routers, including management CPU to ASIC interfaces over I2C/PCIe.
• Experience with modern network operating systems such as SONiC, especially the platform/BSP layer (platform drivers, platform API, sensors, transceivers, thermal/power integration).
• Experience with simulation and emulation environments, including QEMU, hardware behavioral models, or RTL-based SoC simulations for pre-silicon validation.
• Strong debugging skills, including board bring-up logs, kernel crash analysis, JTAG debug, and low-level system trace analysis.