Sr. Staff ML Engineer, FM Training Integration ML Compute

We are looking for a ML Engineer to join our ML Compute team to help improve the efficiency scalability and reliability of model training and inference workloads in the this role you will lead the integration of large-scale ML workloads with cloud infrastructure working cross-functionally with ML engineers infrastructure engineers and researchers to optimize performance improve system efficiency and drive high utilization of accelerator resources.

We are a group of engineers to support training foundation models at Apple! We build infrastructure to support training foundation models with general capabilities such as understanding and generation of text images speech videos and other modalities and apply these models to Apple products. We are looking for engineers who are passionate about building systems that push the frontier of deep learning in terms of scaling efficiency and flexibility and delight millions of users in Apple products.

Own the integration of large-scale model training workloads with accelerator-based cloud infrastructure ensuring scalable and reliable performance optimization across the ML stack including data pipelines model execution and distributed systems to improve throughput latency and hardware and run benchmarks to evaluate model performance and infrastructure configurations using results to guide optimization and improve tooling for observability profiling and debugging to increase visibility and reliability of ML cross-functionally with ML engineers infrastructure engineers and researchers to improve system efficiency and and promote best practices for performance tuning and resource high-quality design and code reviews share best practices and elevate engineering standards across the team.

5 years of experience in software engineering ML infrastructure or related -on experience with machine learning workflows including training evaluation and inference at in Python and experience with at least one major ML framework (e.g. PyTorch or JAX).nExperience with cloud-based infrastructure and distributed systems (e.g. containers orchestration storage and networking).nBachelors degree in Computer Science Engineering or a related field.

Experience working with accelerator-based systems (e.g. GPUs/TPUs) including performance tuning an debugging of ML -on experience with distributed training or inference at scale (e.g. data model or pipeline parallelism).nExperience optimizing large-scale ML systems including bottleneck analysis across compute memory and with profiling tracing and benchmarking tools for ML workloads (e.g. PyTorch Profiler NVIDIA Nsight).nExperience building or operating ML infrastructure using containerization and orchestration frameworks (e.g. Docker Kubernetes).nAdvanced degree in Computer Science Engineering or a related field.