Linux Kernel & Driver Engineer - HPC/AI Fabrics

• Design & Optimize Device Drivers: Develop, maintain, and upstream the open-source hfi1 kernel driver and related subsystems (such as InfiniBand verbs and RDMA core).
• Hardware-Software Co-Design: Partner closely with silicon architects and hardware developers to define register interfaces, MMIO pathways, command queues, and hardware-software contracts.
• Develop Zero-Copy Data Paths: Design and optimize low-latency, high-throughput DMA and RDMA transport engines, minimizing buffer copies and maximizing CPU-bypass capabilities.
• Debug Complex Kernel Concurrency: Identify and resolve intricate kernel-space race conditions, deadlocks, and memory issues under heavy multi-threaded, asynchronous networking workloads.
• Upstream & Community Engagement: Actively submit patches, participate in code reviews, and represent Cornelis within the Linux Kernel Mailing List (LKML) and open-source networking communities.
• Package & Build Automation: Maintain and optimize system build environments, kernel-module packages (DKMS, RPM, Kbuild), and automated integration tests.

• Education: BS, MS, or Ph.D. in Computer Science, Computer Engineering, or a related field (or equivalent practical experience).
• Kernel-Space Mastery: 3+ years of professional experience writing production-grade C code inside the Linux kernel (kernel modules, LKM, memory management, or interrupt handlers).
• High-Speed Networking Protocol Knowledge: Direct experience with RDMA, InfiniBand (IB) Verbs, RoCE, or high-performance user-space bypass frameworks (such as libfabric / OFI or DPDK).
• Hardware Interface Fundamentals: Strong understanding of PCIe architectures, DMA engines, memory mapping (mmap), and MMIO.
• Advanced Kernel Debugging: Hands-on proficiency with kernel analysis tools including KASAN, kmemleak, ftrace, tracepoints, kprobes, and core crash dump analysis.
• Scripting & Automation: Proficiency in scripting languages (e.g., Python, Bash) for automated testing and performance profiling.

Nice to Haves:

• Active track record of contributions to upstream kernel.org (specifically under drivers/infiniband/ or drivers/net/).
• Familiarity with kernel storage protocols (e.g., Lustre, NFS, SRP).
• Experience with GPU-direct communication technologies (e.g., GPUDirect RDMA, DMA-buf).

At Cornelis we're building the future of AI and HPC networking with an AI-first approach to silicon and software development. We're seeking engineers who are energized by working on cutting-edge ASIC design and distributed software systems, and who are motivated to push the boundaries on how AI can transform everything from chip architecture to system performance at scale. Cornelis Networks delivers the world's highest performance scale-out networking solutions for AI and HPC datacenters. Our differentiated architecture seamlessly integrates hardware, software and system level technologies to maximize the efficiency of GPU, CPU and accelerator-based compute clusters at any scale. Our solutions drive breakthroughs in AI & HPC workloads, empowering our customers to push the boundaries of innovation. Backed by top-tier venture capital and strategic investors, we are committed to innovation, performance and scalability - solving the world's most demanding computational challenges with our next-generation networking solutions. We are a fast-growing, forward-thinking team of architects, engineers, and business professionals with a proven track record of building successful products and companies. As a global organization, our team spans multiple U.S. states and six countries, and we continue to expand with exceptional talent in onsite, hybrid, and fully remote roles. Cornelis Networks is seeking a talented Linux Kernel and Driver Developer to architect and optimize our next-generation High-Performance Computing (HPC) and Artificial Intelligence (AI) fabric software stack. Your mission will center on development, optimization, and upstream maintenance of host driver software, focusing on our open-source hfi1 kernel driver and our high-performance user-space Omni-Path Express (OPX) libfabric provider. You will collaborate directly with silicon architects, hardware engineers, and the global open-source community to design software that scales to thousands of nodes with sub-microsecond latency. This role is remote from within the United States.