Meta's Traffic Efficiency team improves the performance, capacity, and reliability of the edge infrastructure connecting people to Meta's products. Edge sites face tight space, power, and hardware constraints, so efficiency directly creates usable capacity and reliability. At Meta's scale, even small efficiency gains can deliver substantial infrastructure savings while protecting product quality as demand grows. We seek an engineering expert to set direction and lead cross-stack performance work spanning high-throughput services, operating systems and kernels, networking and NICs, CPU and memory, and server platforms. The task is to identify high-leverage opportunities, choose the right layer for each solution, and drive safe fleet-wide deployment with measurable impact on latency, throughput, resource efficiency, capacity, reliability, or total cost of ownership.

Qualifications:

Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience 7+ years of software-development experience, or a PhD with 4+ years of relevant experience Track record leading ambiguous, multi-team work from problem framing through large-scale production deployment, with measurable results Deep expertise in at least one area, plus significant hands-on experience in another: high-throughput services, operating systems or kernels, networking or NICs, CPU and memory, or server platforms Experience repeatedly diagnosing and improving systems across two or more of those layers Proficiency in C or C++, a scripting language, and production performance tools such as profiling, tracing, telemetry, resource attribution, or hardware counters Ability to communicate impact and trade-offs and influence teams outside direct ownership Ideal candidates combine several of these; mastery of all is not expected Linux internals, kernel networking, drivers, scheduling, packet processing, programmable networking, or NIC offloads Production-representative benchmarking, server or NIC evaluation, large-scale rollout, systems project contributions, or systemic prevention of incidents CPU, cache, memory locality, NUMA, PCIe, power, thermal, or server platform performance C++ systems optimization involving asynchronous I/O, concurrency, memory allocation, caching, transport, or overload handling High-throughput services such as CDNs, proxies, load balancers, caches, gateways, or request-processing systems