Software Engineering Manager, AI Kernels & Performance Optimization — MTIA Software

From the employer

Meta designs and deploys its own AI systems. MTIA — the Meta Training and Inference Accelerator — is Meta's family of in-house AI accelerator ASICs, running recommendation and ranking workloads in production across Meta's data centers today and expanding into generative AI inference and training as successive silicon generations land. The MTIA Software team is part of the AI & Compute Foundation (ACF) organization within Meta Infrastructure. We build the entire stack a chip vendor would normally supply — compiler and LLVM toolchain, runtime, kernel authoring frameworks and libraries, developer tooling, and deep PyTorch integration — co-designing it with the silicon teams generation over generation. Within that stack, the AI Kernels and Performance Optimization team (KORE) owns the layer where silicon architecture meets arithmetic. Our mission is roofline-level performance and programmability at scale: we design and implement high-performance compute and communication kernels (GEMM, attention variants, normalization, collectives, elementwise and reduction fusions, sparse and quantized paths) against novel architectural features including matrix engines, on-chip reduction fabrics, and software-managed memory hierarchies. We ship kernel libraries with broad PyTorch operator coverage, build C++ and Python kernel authoring frameworks and DSL surfaces, and write production kernels against new architectures long before first silicon — turning hardware proposals into measured roofline evidence while the design can still change. We are seeking a Software Engineering Manager to lead a team of kernel and performance engineers doing this work. You will own the performance of workloads that serve billions of people, drive technical strategy for kernel development and optimization, and build a team that can deliver roofline-level performance on MTIA accelerators. Your team will profile and root-cause performance across the full stack — instruction scheduling, memory hierarchy and DMA behavior, on-chip interconnect, multi-device collectives — and drive fixes to the right layer, whether that is the kernel, compiler, runtime, or hardware. You will partner closely with silicon architecture teams on hardware/software co-design, quantifying the value of proposed features with real kernels and advocating for the changes the software stack needs. This is a hands-on technical leadership role where you will contribute directly to architectural decisions while enabling your team to move with speed and clarity. Lead and manage a team of systems engineers delivering on multiple concurrent projects across ambiguous, high-impact areas of Meta's platform infrastructure Drive roadmap creation and technical strategy for systems-level software, ensuring alignment with broader product and engineering goals Contribute directly to technical work including architecture reviews, design documents, and hands-on coding to maintain credibility and unblock the team Identify and resolve systemic issues affecting team velocity, including on-call load, technical debt, and gaps in tooling or automation Partner with product, design, and data science to ensure engineers contribute across disciplines and deliver outcomes that move business metrics forward Set clear goals and expectations for individual engineers, provide timely and actionable feedback, and proactively address performance gaps Manage staged rollouts and feature delivery using feature flagging and A/B testing frameworks, ensuring high engineering quality and system reliability Recruit, onboard, and develop engineers, building a team structure that minimizes single points of failure and scales with organizational needs Communicate team goals, technical decisions, and progress clearly to cross-functional partners and leadership, including written design proposals and postmortems Participate in org-wide engineering contributions, sharing effective workflows, tooling practices, and systems engineering patterns across the broader organization 4+ years of experience in software engineering with a focus on systems-level software, platform infrastructure, or distributed systems 2+ years of experience managing or leading engineering teams, including setting technical direction and supporting engineer growth Experience driving technical roadmaps and executing against them with measurable outcomes in ambiguous or high-complexity environments Experience identifying and resolving engineering efficiency issues such as on-call burden, technical debt, or gaps in automation Experience collaborating across product, design, and data functions to deliver software that achieves business and user impact Track record of hands-on technical contribution as a manager, including authoring design documents, reviewing system architectures, or writing production code Experience with hardware/software co-design cycles: collaborating with silicon architects, validating architectural features with real kernels, and influencing hardware roadmaps based on software performance analysis Experience working on customized AI accelerators (TPU, AI ASICs, or similar non-GPU architectures) with direct involvement in kernel development and roofline enablement Familiarity with transformer and attention kernel design: FlashAttention-class algorithms, KV-cache management, paged and chunked attention, MoE routing and expert dispatch Experience with distributed execution and collective communication (NCCL/RCCL-class primitives, tensor and expert parallelism, overlapping communication with compute) Demonstrated ongoing AI skill development (e.g., prompt/context engineering, agent orchestration) and staying current with emerging AI technologies Experience managing or leading teams working on high-performance kernel development, accelerator software, HPC, or ML systems performance Hands-on experience with kernel development for novel hardware — including pre-silicon bring-up on simulators/emulators and first-light optimization on new silicon Experience building or contributing to high-performance kernel libraries or frameworks — CUTLASS, cuBLAS, cuDNN, Triton, Helion, oneDNN, Composable Kernel, or comparable internal equivalents Demonstrated ability to integrate AI tools to optimize/redesign workflows and drive measurable impact (e.g., efficiency gains, quality improvements) Experience with compiler and codegen technologies relevant to kernels: MLIR, LLVM, TVM, XLA, Halide, or polyhedral scheduling Experience with low-precision numerics and quantization — FP8/E4M3/E5M2, MX and other block-scaled formats, INT8/INT4 — including error analysis and calibration Experience adhering to and implementing responsible, ethical AI practices (e.g., risk assessment, bias mitigation, quality and accuracy reviews)