Senior ML Infrastructure Engineer

Apple • Cupertino, CA

1d

About The Position

The role is on Apple’s AI Platform team, which provides Apple’s ML engineers and researchers with data systems and large-scale compute for building and shipping models. The team owns the data layer that supports large-scale model training, including data ingestion, versioning, lineage, governance, and high-throughput data delivery to training fleets.

The engineer will set technical direction for the platform and own its most challenging system-level problems. This includes defining architecture for ingestion, immutable versioning, lineage, and governance across structured, unstructured, and multimodal data at petabyte scale, so model runs can be reproduced from versioned datasets. The role also defines data-access and loading architecture for large GPU and TPU fleets, along with platform-wide choices concerning formats, dataset abstractions, SDKs, and core libraries.

The position drives technical direction across platform and partner teams, establishes their interfaces and contracts, mentors senior engineers, leads design reviews, and serves as an escalation point for difficult problems. The engineer partners with research and product leadership on the platform roadmap for foundation models, multimodal data, and retrieval-augmented systems, and drives efficiency, reliability, and automation across the data and control planes.

Requirements

• At least 10 years of work experience in machine learning infrastructure, distributed data systems, or a related field.
• At least 10 years of experience building and shipping large-scale data or ML infrastructure and platforms in production.
• Extensive experience architecting and delivering large-scale distributed data or ML infrastructure that multiple teams or products depend on in production.
• A track record of setting technical direction and driving it to delivery across teams.
• Deep systems engineering skills, including strong Python and a systems language; Rust is strongly preferred, with C++ or Go acceptable.
• Hands-on performance engineering experience for I/O-bound workloads, including technologies or approaches such as Arrow, zero-copy, memory mapping, async I/O, and high-throughput object storage.
• Deep familiarity with columnar and lakehouse formats such as Parquet, Iceberg, Delta, or Lance, and the judgment to choose between them at scale.
• Strong working knowledge of end-to-end ML workflows and how training and inference consume data.
• Familiarity with modern ML and generative techniques, including transformers, diffusion, retrieval-augmented generation, and fine-tuning, sufficient to design systems for those workloads.
• Ability to design highly available, easy-to-use systems and mentor and elevate other engineers.
• Strong collaboration and communication skills, including the ability to align multiple teams around a technical direction.
• B.S., M.S., Ph.D. in Computer Science or Computer Engineering, or equivalent practical experience.

Nice To Haves

• Experience defining data or ML platform architecture adopted across an organization.
• Deep experience with the data-loading and dataset-access layer of a modern ML framework such as PyTorch, JAX, or TensorFlow.
• Experience with distributed data-loading frameworks for ML, such as Ray Data, NVIDIA DALI, WebDataset, or Mosaic StreamingDataset.
• Experience feeding data to GPU or TPU fleets at scale and keeping them saturated.
• Experience with data lineage and governance systems such as DataHub, OpenLineage, Unity Catalog, or equivalent.
• Contributions to or operational experience with Spark, Daft, Polars, or DuckDB internals.
• Experience with containerization and orchestration, including Docker and Kubernetes.

Responsibilities

• Own the architecture of the platform behind Apple’s largest model builds, including ingestion, immutable versioning, lineage, and governance across structured, unstructured, and multimodal data at petabyte scale.
• Define data-access and loading architecture for high-throughput delivery to large GPU and TPU fleets.
• Make platform-wide system and format decisions, including columnar and lakehouse strategy, dataset abstractions, SDK design, and core libraries, supported by design and proof.
• Drive technical direction and influence across the platform and partner teams, and define interfaces and contracts between them.
• Mentor senior engineers, lead design reviews, raise the technical bar, and act as an escalation point for difficult system problems.
• Partner with research and product leadership to shape the platform roadmap for foundation models, multimodal data, and retrieval-augmented systems.
• Drive efficiency, reliability, and automation across the data plane and control plane supporting Apple’s ML fleet.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior