Software Engineer, Systems ML - Compilers / Kernels

Meta is seeking a Software Engineer to work on systems-level machine learning infrastructure that powers some of the world's largest AI workloads. In this role, you will design, build, and optimize the software systems that enable efficient training and inference of large-scale deep learning models across Meta's production infrastructure. You will work at the intersection of ML frameworks, distributed systems, and hardware acceleration, contributing to the reliability, performance, and scalability of ML platforms that underpin Meta's AI-driven products across social media, augmented and virtual reality, and generative AI.

Responsibilities

Design and implement scalable systems-level software components that support large-scale ML training and inference workloads across distributed compute infrastructure Identify and resolve performance bottlenecks in ML pipelines through profiling, instrumentation, and targeted optimization of compute, memory, and I/O utilization Develop and maintain high-quality, well-tested software for ML runtime systems, including scheduling, resource management, and execution frameworks Collaborate with AI researchers and hardware teams to co-design software solutions that efficiently map deep learning model architectures onto accelerated hardware platforms Own the technical design of systems components, evaluating trade-offs across correctness, performance, reliability, and maintainability Establish and monitor service-level objectives and performance metrics for ML infrastructure systems, building dashboards and alerting to reduce mean time to mitigation during incidents Drive adoption of engineering best practices including automated testing, staged rollout strategies using feature flagging and A/B testing frameworks, and comprehensive documentation Mentor other engineers on the team by identifying growth areas, sharing technical expertise, and contributing to code and design reviews Partner with cross-functional stakeholders including product, data science, and operations teams to translate ML research requirements into production-ready systems Proactively identify opportunities to improve developer tooling, automation, and engineering efficiency across the ML systems organization

Qualifications

Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience 6+ years of experience in software engineering with a focus on systems software, ML infrastructure, or distributed computing Experience designing and optimizing software systems for large-scale machine learning training or inference workloads Experience with performance analysis and optimization techniques including profiling, bottleneck identification, and low-level systems tuning Experience building reliable, production-grade distributed systems with well-defined observability, monitoring, and incident response practices Experience communicating technical decisions through design documents, architectural proposals, or cross-functional stakeholder alignment Experience with ML frameworks such as PyTorch, TensorFlow, or JAX, including framework internals, operator development, or runtime optimization Demonstrated ongoing AI skill development (e.g., prompt/context engineering, agent orchestration) and staying current with emerging AI technologies Demonstrated ability to integrate AI tools to optimize/redesign workflows and drive measurable impact (e.g., efficiency gains, quality improvements) Experience with compiler technologies or graph-level optimizations for deep learning, such as MLIR, LLVM, TVM, or XLA Experience with GPU or AI accelerator programming models such as CUDA, ROCm, or vendor-specific kernel libraries Experience adhering to and implementing responsible, ethical AI practices (e.g., risk assessment, bias mitigation, quality and accuracy reviews) Experience applying responsible AI practices in systems design, including bias mitigation, model quality evaluation, or AI risk assessment