Responsibilities

• Define and standardize AI-chip hardware interfaces while designing scalable, reusable, and maintainable driver frameworks.
• Develop, port, and optimize drivers for heterogeneous computing components, including NPUs, ISPs, VPUs, 2D engines, and DMA controllers.
• Build, maintain, and enhance the AI-chip BSP, covering bootloader and startup sequences, clock and reset management, GIC interrupt control, and PMIC/power-management subsystems.
• Develop and support peripheral drivers for interfaces such as PCIe, Ethernet (AVB/TSN), CAN/CAN FD, UART, I2C, SPI, GPIO, and USB.
• Ensure reliable and efficient data exchange across multi-sensor platforms and system-level communication interfaces.
• Work closely with hardware, firmware, system software, and validation teams to troubleshoot issues, optimize performance, and deliver production-grade driver solutions.

Requirements

• Bachelor's degree or higher in Electronics, Computer Science, Software Engineering, Automation, or a related field.
• Minimum 3 years of chip or embedded driver development experience across one or more domains, including low-speed and high-speed interfaces, imaging and multimedia systems, platform software, security modules, bootloaders, or AUTOSAR Classic Platform (CP).
• Strong proficiency in C/C++ programming, with solid knowledge of ARM architecture, assembly language, and common debugging and profiling tools.
• Deep understanding of Linux kernel driver development, including Platform Drivers, Device Tree architecture, Kernel Modules, as well as memory management, interrupt handling, DMA operations, and multi-threading/concurrency mechanisms.
• Experience in virtualization driver development is highly desirable.
• Strong hands-on experience with AUTOSAR development, particularly AUTOSAR Classic Platform, is preferred.
• Excellent debugging, problem-solving, and cross-functional collaboration skills, with the ability to work effectively across hardware and software domains throughout the product development lifecycle.