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.

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