Linux Kernel & Device Drivers

Location: Samsung Semiconductor San Jose CA

Experience: 5 to 15 years (depending on grade)

As a Senior Engineer in the Systems Software team you will drive the software-hardware co-design for Samsungs AI and data center solutions. This role requires a visionary approach to Linux Kernel Memory Management specifically focused on heterogeneous memory virtualization and high-bandwidth interconnects for our next-generation SoC and SSD platforms.

• Memory Management R&D: Architect and optimize Linux kernel memory management for heterogeneous systems including UVM (Unified Virtual Memory) memory tiering and CXL-based memory expansion.
• Kernel & Driver Design: Lead the design of Linux device drivers for high-performance interfaces such as PCIe Gen5/6 NVMe and proprietary AI accelerators.
• Virtualization & Hypervisors: Develop and tune KVM and QEMU support for IOMMU interrupt virtualization and hardware-assisted memory management.
• SoC Bring-up & Architecture: Partner with hardware architects to define registers and memory maps for upcoming ARMv9 and RISC-V silicon.
• System-Level Debugging: Resolve critical system bottlenecks and memory corruption issues using advanced tools like Lauterbach TRACE32 hardware emulators (Palladium/Zebu) and kernel profilers.

• Education: MS in Computer Science Computer Engineering or a related field.
• Kernel Deep-Dive: Expert knowledge of the Linux MM subsystem (paging swapping HugePages page cache and LRU eviction policies).
• Hardware Interface Mastery: Deep understanding of PCIe/CXL protocol stacks cache coherency (AMBA CHI/ACE) and DMA engines.
• Low-Level Programming: Expert proficiency in C and Assembly (ARM/x86); experience with Python for automation and performance modeling.
• Security & Reliability: Familiarity with hardware security features like TrustZone ARM CCA and memory protection units.

• Significant contributions to the mainline Linux Kernel (specifically in the mm/ or drivers/pci/ directories).
• Experience with Cloud and Data Center workloads and understanding their impact on kernel scheduling and memory latency.
• Knowledge of Machine Learning frameworks and how they interact with kernel-level memory allocators.