Meta is investing heavily on the Meta Cloud. As part of this push we want to build a cloud-native infrastructure optimized for AI development. The Host Networking team owns the networking stack that runs on the compute platforms at Meta, spanning the software dataplane/host networking platform and select custom hardware components that power connectivity, performance, and security across our fleet. As part of the Meta Compute Initiative, we’re evolving this foundation to support a cloud-native, Kubernetes-first AI infrastructure—including platform primitives like network virtualization, load balancing/service connectivity, and distributed firewall/policy enforcement—at Meta scale. This is a chance to lead architecture and delivery while influencing a broad cross-functional ecosystem, leveraging (and contributing to) open source where it helps us move faster. We build the platform layer and partner closely with the many application teams that run on top.
Responsibilities
Architect and lead delivery of Meta Compute’s cloud-native host networking platform, owning end-to-end technical direction from design through production rollout Build/modernize Kubernetes networking features integrated with Meta’s host networking stack (e.g., dataplane integration, operability, scaling characteristics) Design and implement core networking primitives: network virtualization, load balancing/service connectivity, and distributed firewalling/policy enforcement Drive time-to-market execution: define milestones, de-risk key technical choices early and deliver iterative production increments Lead broad cross-functional alignment across compute, storage, platform, security, and hardware/fabric partners—owning interfaces/contracts and end-to-end outcomes Leverage and influence open-source ecosystems: evaluate open-source building blocks, contribute improvements where needed, and ensure Meta’s solution is maintainable and operable long-term
Qualifications
Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience 10+ years software development experience in industry settings or PhD with 4+ years of experience 4+ Years Proven experience designing and shipping networking dataplane/control-plane systems in production (e.g., host networking, L4/L7 load balancing, NAT/conntrack, firewalls/policy, network virtualization) Hands-on Kubernetes networking experience in production (e.g., CNI, Services, NetworkPolicy/policy models, service connectivity/load balancing integration) Proficiency in C/C++ and at least one scripting language (Python/Shell Scripting) Experience with developing and automating test suites Demonstrated experience working across disciplines to align on technical decisions and deliver integrated solutions Experience building or contributing to Kubernetes operators, controllers, or device plugins that manage hardware-accelerated networking or storage resources Direct experience with DPU SoC/SmartNIC Platforms Familiarity with Linux networking internals such as netdev subsystem, TC/iptables/nftables, network namespace isolation, traffic shaping, and kernel driver development Experience in hyperscale or cloud-provider environments Expertise in high-performance networking frameworks and techniques such as DPDK, eBPF/XDP, AF_XDP, io_uring, kernel bypass, zero-copy I/O, and RDMA/RoCEv2 Experience collaborating closely with hardware and ASIC design teams, contributing to hardware specifications, co-designing hardware-software interfaces, and influencing silicon feature sets based on software requirements