Senior Linux Kernel / ROCm Developer

Job Description: Senior Linux Kernel / ROCm Developer

Job Title: Senior Linux Kernel / ROCm Developer

Experience: 7+ Years

Domain: Linux Kernel, GPU Drivers, ROCm and GPU Compute

Qualification: Bachelor's or Master's degree in Computer Science, Computer Engineering, Electrical Engineering, or a related technical discipline

Role Overview

We are looking for a self-motivated and technically strong Senior Linux Kernel / ROCm Developer with hands-on expertise in Linux kernel development, device drivers, GPU compute and the ROCm software stack.

The engineer will be responsible for developing, debugging, optimizing and maintaining kernel-level and ROCm components for AMD GPU platforms. This role requires close collaboration with hardware, firmware, compiler, runtime, validation and application teams to resolve complex issues across software and hardware boundaries.

The ideal candidate should be capable of independently driving technical investigations from initial analysis through root-cause identification, corrective action and closure.

Key Responsibilities

• Design, develop, debug and maintain Linux kernel, GPU driver, KFD, DRM, amdgpu and ROCm-related components for AMD GPU platforms.
• Analyze, reproduce, root-cause and resolve complex issues involving the Linux kernel, GPU drivers, ROCm runtime and GPU compute workloads.
• Support GPU platform bring-up and enablement in collaboration with hardware, firmware, compiler, runtime, validation and application engineering teams.
• Debug issues across kernel, driver, firmware, runtime and hardware layers.
• Use advanced Linux and GPU debugging tools, including:kgdb
• ftrace and tracepoints
• perf
• dmesg
• Crash dumps
• ROCm profiling tools
• GPU diagnostic tools
• Conduct performance profiling, bottleneck analysis and optimization of kernel, driver, runtime and GPU compute workloads.
• Improve platform stability, scalability, efficiency and overall GPU compute performance.
• Independently lead technical investigations and recommend appropriate design or architectural improvements.
• Participate in architecture discussions, design reviews, code reviews and defect triage.
• Develop high-quality, maintainable code in line with Linux kernel and open-source development practices.
• Collaborate effectively with geographically distributed engineering teams.
• Stay current with developments in Linux kernel technology, GPU architecture, ROCm, heterogeneous computing and open-source software.

Required Skills and Qualifications

• Bachelor's or Master's degree in Computer Science, Computer Engineering, Electrical Engineering, or a related field.
• Minimum 7+ years of relevant industry experience in one or more of the following areas:Linux kernel development
• Linux device-driver development
• GPU compute software
• ROCm software development
• Low-level system software
• Strong understanding of Linux kernel architecture and internals, including:Memory management
• Process and thread scheduling
• Interrupt handling
• Synchronization and concurrency
• Kernel modules
• DMA and device interaction
• Power management
• Strong hands-on experience in Linux device-driver development and debugging.
• Familiarity with AMD GPU architecture, KFD, ROCm software stack, heterogeneous compute or similar accelerator technologies.
• Strong programming expertise in C and C++.
• Proven experience debugging complex Linux-based system software issues.
• Experience in system-level and kernel-level performance profiling, bottleneck analysis and optimization.
• Good understanding of interactions between software, firmware and hardware.
• Ability to independently manage technical tasks and drive issues to completion with minimal guidance.
• Strong analytical, debugging and root-cause analysis skills.
• Excellent ownership, accountability and execution focus.
• Strong verbal and written communication skills.
• Ability to work effectively with cross-functional and globally distributed teams.

Preferred Skills

• Hands-on experience with one or more of the following:KFD
• DRM subsystem
• amdgpu driver
• ROCm runtime
• HIP
• HSA
• GPU compute-driver development
• Experience contributing to the upstream Linux kernel, open-source GPU drivers, ROCm components or other community-driven software projects.
• Knowledge of GPU memory management and heterogeneous memory-management concepts.
• Experience with GPU virtualization or accelerator virtualization.
• Understanding of:PCIe
• IOMMU
• NUMA
• DMA
• Interrupt handling
• Power and thermal management
• Experience with GPU or server-platform bring-up.
• Familiarity with GPU profiling, diagnostics and performance-analysis tools.
• Exposure to compiler, runtime or heterogeneous computing frameworks.