Applied Machine Learning Engineer, Platform Architecture

Join the SoC Architecture team building ML and generative AI systems that shape how future Apple silicon is architected and tuned. Were looking for an AI/ML Engineer who can turn complex hardware data into architectural insight. You will apply that expertise to studying and improving the performance and power behavior of modern System-on-Chip designs across the full product lifecycle: ML-driven research to identify what should change in hardware or software hands-on partnership with silicon and OS teams to implement and bring those changes up on real silicon and seeing them through to ship. This role is ideal for a hands-on ML engineer who is energized by both research and shipping product and who thrives at the intersection of large-scale data system architecture and ML.

You will join a multidisciplinary team of ML software and architecture engineers building systems that drive architectural exploration and tuning for current and future Apple SoCs. The work targets the full performance-power tradeoff space across fabric memory subsystem system caches dynamic voltage and frequency state control clock and power gating policies sleep state controls and bottleneck prevention.

Design and build advanced ML and generative AI workflows to automate workload analysis and SoC architectural modern deep learning and generative AI techniques to large-scale workload data including the time-series telemetry that is central to this work to inform software and hardware architectural insights into concrete hardware and software changes that improve performance power and performance per with hardware software and architecture teams across Apple to implement those changes validate them on real silicon and see them through to shipping product.

B.S. in Computer Science Computer Engineering Electrical Engineering or a related ML industry experience deploying complex ML systems in applying modern ML techniques to large real-world experience in Python and experience in modern deep learning frameworks.

Working knowledge of SoC compute memory and power-management subsystems how real workloads exercise them and the C/C modeling infrastructure typical of SoC in time-series analysis including feature engineering on streaming applying ML beyond prediction including driving decisions optimizing policies and efficiently searching large configuration record of training large-scale models across distributed building stateful multi-turn agentic frameworks and complex execution .S. or Ph.D. in Computer Science Computer Engineering Electrical Engineering or a related field and 10 years of relevant record of moving quickly from hypothesis to result and iterating based on communication and collaboration skills to work effectively across technical disciplines.