Staff Software Engineer - Linux Kernel

Fastly•Quinte West, ON

8h•$211,370 - $253,644•Remote

About The Position

Fastly helps people stay better connected with the things they love. Fastly’s edge cloud platform enables customers to create great digital experiences quickly, securely, and reliably by processing, serving, and securing our customers’ applications as close to their end-users as possible — at the edge of the Internet. The platform is designed to take advantage of the modern internet, to be programmable, and to support agile software development. Fastly’s customers include many of the world’s prominent companies, including GitHub, Yelp, Paramount, and JetBlue. We're building a more trustworthy Internet. Come join us. Fastly's Kernel and Host Networking team builds and ships the Linux kernel that runs on every server in our global edge network, and every customer request passes through it. As a Staff engineer you own technical problems end to end: from a crash signature or a new NIC on the bench, through root cause and design, to a change that is upstreamed and safely running on thousands of servers. The work spans networking, XDP and eBPF, memory, performance and hardware enablement. It also spans layers. Our kernel is the datapath of a distributed system whose control planes are constantly pushing new state to the fleet, so staying correct under change is part of the job. Formal CS education is not required; demonstrable experience is what matters.

Requirements

7+ years of related engineering experience, including substantial production C, with 3 of those years leading technical projects
Production Linux kernel development, including debugging use-after-free, refcount and leak bugs with KASAN, kdump, ftrace, perf or bpftrace
Working knowledge of the scheduler and memory management (EEVDF, memcg)
Experience managing a kernel patch stack and working with upstream communities
Solid understanding of the Linux network stack (TCP/IP, sk_buff lifecycle, qdiscs, NIC drivers) and XDP or eBPF, with deep expertise in at least one of these areas
Operational discipline: staged rollouts, canaries, kill switches, observability, postmortems and closing action items
Understanding of distributed systems: control plane and datapath consistency, partial failure and safe config rollout at fleet scale
Ability to design broadly scoped work and drive it across teams

Nice To Haves

Upstream kernel contributions or subsystem maintainership
Deep Knowledge of Transport protocols from IETF RFCs: QUIC, TCP, ECN and L4S, MASQUE
Hardware enablement on AMD EPYC or Intel platforms, and NIC drivers such as mlx5, ice or i40e
Load balancing or traffic steering systems
CDN or large edge network environments
Container networking: network namespaces, veth, cgroups, and eBPF-based CNI datapaths

Responsibilities

Set technical direction for kernel projects, own them end to end from problem statement to fleet rollout, and mentor engineers on the team
Take new servers, CPUs and NICs from first boot to full fleet: drivers, IRQ affinity, RSS and queue layout on high core count hosts
Improve datapath and transport performance with XDP, eBPF and protocol work based on IETF specifications
Debug problems that cross layers with application teams, such as MTU behavior, cgroup memory accounting and packet filtering
Develop and maintain kernel changes and Fastly kernel modules on our long term kernel line, and reduce our private delta by upstreaming
Make sure state pushed from control planes to eBPF and XDP programs is applied, verified and safe to roll back
Own kernel health in production: triage crashes, WARNs and regressions across the fleet, find the root cause and ship the fix
Own and improve the kernel release process and the signals behind it: CI, bake, canary, staged rollout and rollback across thousands of servers, plus crash metrics and drift detection that fire before customers notice
Join the on-call rotation, lead the kernel side of incidents and drive follow-up actions to closure