Systems Engineer, Kernel (Networking)

Senior Systems Engineer, Kernel Networking

CoreWeave is seeking a specialized Kernel Networking Engineer to join our HAVOCK Team. In this role, you will be the subject matter expert for the networking subsystem of CoreWeave’s Linux-based infrastructure. As we scale our massive AI/HPC clusters, you will focus on optimizing the datapath, tuning the TCP/IP and RDMA stacks, and ensuring the stability of high-throughput workloads across NVIDIA, Mellanox, and Broadcom hardware.

Hardware - Acceleration - Virtualization - Operating Systems - Containerization - Kubelet

Our Team’s Stack:

• Python, Go, bash/sh, C
• Custom Linux Kernel, Ubuntu
• Debug Tools: crash, kdump, drgn, gdb
• Prometheus, Victoria Metrics, Grafana, Loki
• Docker, kubernetes (k8s), KubeVirt

Focus Areas:

• Holistic Troubleshooting – Act as the first line of defense for complex system crashes, soft lockups, and kernel panics.
• Cross-Domain Debugging – Identify whether a root cause lies in memory management, storage, or the network layer.
• Incident Response – Reduce "Mean-Time-To-Resolution" by quickly analyzing crash dumps and stack traces.
• Reliability Engineering – Contribute to the "Smarter Triaging" initiative to automate crash analysis.
• Fleet Stability – Ensure kernel support across diverse hardware (CPUs, GPUs, DPUs).

Responsibilities:

• Analyze kernel crashes, oopses, and panics across the entire stack.
• Apply specific networking knowledge to troubleshoot issues with NVIDIA/Mellanox/Broadcom NICs.
• Utilize crash dump analysis (kdump, crash, drgn) to triage issues affecting customer workloads.
• Improve documentation and RCA processes for kernel failures.
• Assist in maintaining kernel builds and CI/CD pipelines to streamline testing.

Requirements:

• 5+ years of experience in systems-level development or kernel engineering.
• Broad Kernel Knowledge: Solid grasp of memory management, scheduling, and filesystems.
• Networking Fluency: Proven record troubleshooting RoCE, IB, and RDMA issues.
• Debugging Mastery: Expert capability with standard utilities and a systematic approach to root-cause analysis.
• Excellent verbal and written communication skills (ability to explain complex kernel bugs to stakeholders).

Nice-to-haves:

• Experience with eBPF for troubleshooting.
• Knowledge of GPU/NVLink architectures.
• Experience working with automated monitoring/alerting systems (Grafana, Jira automation).
• Willingness to present at conferences (LPC, LSFMMBPF).

Our compensation reflec