Systems Engineer, Kernel

Weights & Biases Weights & Biases · Data AI · Bellevue, WA +2 · Technology

CoreWeave is seeking a Systems Kernel Engineer to join their HAVOCK Team. The role focuses on maintaining and improving the stability, performance, and evolution of CoreWeave’s Linux-based infrastructure, with responsibilities including debugging kernel-level issues, analyzing crashes, and upstreaming fixes and features. The ideal candidate will have deep experience in low-level systems engineering and understand how modern workloads stress kernels, working across CPUs, GPUs, DPUs, networking, and storage.

What you'd actually do

  1. Analyze kernel crashes, oopses, panics, and dumps to identify root causes and propose fixes.
  2. Develop patches for the Linux kernel and upstream them where applicable (networking, storage, virtualization, GPU/DPU enablement).
  3. Support new hardware bring-up across Intel, AMD, ARM CPUs, NVIDIA GPUs, DPUs, and NICs.
  4. Tune kernel subsystems for latency, throughput, and scalability in distributed HPC/AI clusters.
  5. Implement diagnostics and tooling for kernel-level observability.

Skills

Required

  • 5+ years of professional experience in Linux kernel engineering or systems-level development
  • Deep understanding of kernel internals (memory management, scheduling, networking, storage, drivers)
  • Experience debugging kernel crashes, dumps, and panics using tools like crash, gdb, kdump
  • Strong C programming skills
  • Experience working with kernel modules, drivers, and subsystems
  • Strong problem-solving abilities with a “full-stack” systems perspective

Nice to have

  • Contributions to the Linux kernel or related open-source projects
  • Familiarity with virtualization (KVM, QEMU, VFIO) and container runtimes
  • Networking stack expertise (InfiniBand, RoCE, TCP/IP performance tuning)
  • GPU/DPU bring-up and driver experience
  • Experience in HPC or large-scale distributed systems
  • Familiarity with QA/QE best practices
  • Experience working in Cloud environments
  • Experience as a software engineer writing large-scale applications
  • Experience with machine learning is a huge bonus

What the JD emphasized

  • kernel-level issues
  • kernel crashes
  • kernel panics
  • kernel dumps
  • upstream fixes
  • kernel generalist
  • low-level systems engineering
  • modern workloads stress kernels
  • diverse hardware/software ecosystem
  • Kernel Debugging
  • Upstream Contributions
  • Stack-Wide Support
  • Kernel-Hardware Enablement
  • Performance & Stability
  • kernel crashes
  • performance regressions
  • upstream kernel patches
  • kernel-level observability
  • kernel readiness
  • kernel-level expertise
  • Linux kernel engineering
  • kernel internals
  • kernel crashes
  • kernel dumps
  • kernel panics
  • kernel modules
  • kernel subsystems